Elsevier

Safety Science

Volume 48, Issue 7, August 2010, Pages 859-867
Safety Science

The context and habits of accident investigation practices: A study of 108 Swedish investigators

https://doi.org/10.1016/j.ssci.2010.04.001Get rights and content

Abstract

The context and habits of accident investigation practices were explored by means of questionnaire data obtained from accident investigators in the healthcare, transportation, nuclear and rescue sectors in Sweden. Issues explored included; resources, training, time spent in different phases of an investigation, methods and procedures, beliefs about causes to accidents, communication issues, etc. Examples of findings were: differences in the extent to which the ‘human factor’ was perceived as a dominant cause to accidents; manning resources to support investigations were perceived as rather scarce; underutilization of data from safety related processes such as risk analysis and auditing data; the phase of suggesting remedial actions (recommendations) were comparatively brief and generally not well supported. A majority of the investigators thought that the investigations were free from pressures to follow a specific direction; the investigators also thought that performing an investigation in itself (regardless of the specific results) had positive influences on safety. A majority of the investigators thought that upper management had a relatively strong influence on safety in the organizations. The results are discussed in terms of suggestions for strategies to strengthen investigation practices – particularly those conducted as part-time work in organizations.

Introduction

Accident and incident investigations are vital ingredients of safety management. The academic interest in accident investigations has traditionally focused on results from specific investigations, typically major accidents (Balasubramanian and Louvar, 2002, Drogaris, 1993, Le Coze, 2008), and on various methods and techniques for experience feedback including accident investigation (Benner, 1975, Dien et al., 2004, Johnson, 2002, Johnson and Holloway, 2003, Kjellén, 2000, Katsakiori et al., 2009, Sklet, 2004). Little attention is normally devoted to the organizational context in which an investigation takes place. Consequently, we have few answers to questions such as: how much time is spent on the different phases of an investigation (planning, analysis, etc.); how much and what type of training support investigator practices; which professional networks (if any) support investigators practices? Answers to questions such as these may be useful to make the process of accident investigation more efficient, reliable, and precise.

Another important set of questions concern investigators’ personal beliefs about accident causation and how these might affect investigation practices. For example, how are general concepts such as ‘the human factor’ and ‘safety culture’ perceived and applied in investigation practices, and how do accident investigators think about ‘causes’ to events in those branches in which they operate?

The present study (conducted in 2008) aimed at providing answers to questions such as those above with the help of a sample of Swedish accident investigators who work in different sectors (health care, nuclear, rescue service, etc.). The study is part of an ongoing research program that aims to shed light on different factors that affect accident analysis and design of remedial actions. The results will be used to support accident investigator practices by means of training, seminars, etc.

A questionnaire was used comprising both open questions to be answered in free text and evaluation scales were subjects had to rate questions about accident investigation practices. Data from 108 accident investigators from various branches are included in this study. At present, our research has been mainly descriptive and hypothesis generating. The results and statistics presented below should be interpreted with care since the sample is limited. However, we feel that some of the observations are interesting enough to be presented to a wider audience, in the hope that the results can serve as input and inspiration for further and more extensive research.

Section snippets

The questionnaire

A questionnaire was developed focusing on the procedures, context, beliefs, etc., that are associated with accident investigation practices. The questions were partly based on a tentative generic model about different phases in accident investigation, used in a previous review of mainly Swedish, accident manuals (Lundberg et al., 2009). The present questionnaire included the following themes:

  • Investigators characteristics and background: age, gender, affiliation, role in organization, percentage

Efforts and resources

Table 3 shows the responses to the question about how many investigators normally participated in an investigation, while Table 4 shows the average duration of an investigation.

Among those who performed an investigation themselves, the transportation and rescue samples dominated with 44%, respective 34%. In the nuclear group, there was always more than one investigator and the same response was obtained for the patient safety sample.

Table 4 shows that the nuclear sector in average spent most

Discussion

To our knowledge, data of the type presented here have not been available to any greater extent before. Roed-Larsen et al. (2004) has presented some data that addressed partly similar issues. The responses given in that survey were, however, not seen as representative of the population but rather regarded as ‘a kind of snapshot’ (the questionnaire was distributed to 150 institutions/organizations and response rate was about 30%). The findings obtained in the present study should also be

Acknowledgement

This research was sponsored by the Swedish Civil Contingencies Agency, through the project “Assumptions about accidents and their consequences for investigation and design of remedial actions.”

References (17)

There are more references available in the full text version of this article.

Cited by (41)

  • Planning and implementing remedial measures from incident investigations: A study of the Ghanaian mining industry

    2020, Safety Science
    Citation Excerpt :

    In addition, many organisations lack the ability to effectively learn from past incidents due to the gap between the identification and implementation of lessons (Hopkins, 2008; Kjellén, 2000; Kletz, 2001). There have been recent calls to increase the focus on the process of improvement that follows the identification of causes (Carroll & Fahlbruch, 2011; Drupsteen & Hasle, 2014; Lundberg et al., 2009, 2010; Rollenhagen et al., 2010). In this study, we examined the planning and implementation of recommendations of incident investigations in the Ghanaian mining industry.

View all citing articles on Scopus
View full text