Incident reporting or storytelling? Competing schemes in a safety-critical and hazardous work setting

Abstract

Incident-reporting schemes can prevent accidents through organizational learning from incidents. However, many occupational health and safety incidents go unreported. For these reasons I undertook ethnographic fieldwork to investigate the low level of reporting among railway maintenance technicians in Sweden and the role played by informal storytelling within their occupational communities. The study found that the incident-reporting scheme is not integrated in technicians’ practices and cultural frame and does not seem to serve their interests. Storytelling, however, is an integral part of technicians’ practices and their accident etiology and creates a way for them to address risks, at least from a narrow perspective. The occupational etiology is based upon technicians’ local practice, which emphasizes vigilance, carefulness, skill, responsibility, and the like, and usually neglects root causes. This frame is rational and intelligible, given the technicians’ limited power to influence their working conditions, as well as their limited training and the poor feedback they receive when incidents are reported. However, the occupationally-based perspective impedes the articulation of a systems perspective that could be used for organizational learning. To make an incident-reporting scheme work, employees must be given ownership, must know how and why to use it, and need feedback on root causes. These root causes must also be addressed.

Introduction

Railway technicians are responsible for maintaining the physical infrastructure of the railway to support safe and timely transport. Their work necessarily exposes them to a variety of physical dangers such as trains running along the tracks and high-voltage power lines. Incident-reporting schemes are therefore extremely important in order to prevent accidents. They are also important in occupations where similar situations arise, such as police work, fire-fighting, and nursing. Unfortunately, in the case studied for this paper, the number of reported occupational health and safety incidents was very low, impeding the usefulness of the incident-reporting scheme. This article seeks to explain the reasons for the low number of reported incidents by comparing the characteristics and functioning of the incident-reporting scheme from the technicians’ perspective with that of informal storytelling. In addition, the consequences of low reporting for organizational learning from incidents and for systematic occupational health and safety management are discussed.

In 2002, I attended a workplace meeting at Banverket Produktion, which is the major contractor for railway maintenance in Sweden. One of the supervisors told the attending technicians that the number of reported occupational health and safety incidents was lower than the number of accidents, which he thought was strange: “This cannot really be the case, there must be more incidents. Don’t feel ashamed to report them”. In response though, one of the technicians argued that “it is difficult to know what constitutes an incident if nothing happens” (Fieldwork notes, May 2002). The supervisor’s comment suggested both that he was influenced by ideas of the iceberg metaphor but also that he mainly attributed technicians reluctance to report incidents because they felt they had caused the incidents through shameful practice. The technician’s response though, suggested that technicians might have a different conceptualization of incidents.

In 2003, in response to corporate and regulatory demands for a systematic safety management system, Banverket Produktion introduced the safety management system Synergi, replacing previous incident-reporting schemes (Järnvägsinspektionen, 2002). Banverket Produktion anticipated success because of the system’s technical sophistication and its success in the Norwegian off-shore activities for which it had been developed.¹ However, while the number of reports of transport-related incidents has increased, the number of reported occupational health and safety incidents has not.² This article argues that the official incident-reporting scheme in Banverket Produktion has been outperformed by technicians’ storytelling in terms of the number of reported occupational health and safety incidents.

There are many reasons why storytelling is the preferred mode. Different accident etiologies shape what is considered an incident in different communities. They also shape the pattern of reporting and non-reporting, as well as what are considered appropriate measures after an incident. For one thing, the incident-reporting scheme is not integrated into technicians’ practices and cultural frame and it does not seem to serve their interests. Storytelling, however, is an integral part of their practices and their specific accident etiology and it provides a way for the technicians to address risks, although, as I will point out later, from a narrow perspective. The technicians’ accident etiology gives meaning to their work, to their understanding of accidents, and to the “repair” work required. The occupational etiology is based upon their local practice, emphasizing vigilance, carefulness, skill, responsibility and so forth, and usually neglects root causes. This frame is locally rational and intelligible, given the technicians’ limited power to influence their working conditions, as well as their limited training and the poor feedback they receive when incidents are reported.

However, storytelling is of limited value from the perspective of organizational learning and occupational protection. Firstly, stories are not as widely shared as they warrant, which limits the opportunity to learn how to prevent similar events to the circles in which the stories are told. Secondly, stories are shaped by the shared values and norms within the social context in which they are told. Thus technicians emphasize attention, vigilance, personal responsibility, carefulness and the like as the major means to maintain safe practice, but pay too little attention to the wider context of accident causation. Consequently, storytelling is not a perfect substitute for incident-reporting systems. Possible ways of addressing these drawbacks of storytelling are discussed in relation to each of the three incidents presented here and in the conclusions.

Three major theoretical perspectives underlie the arguments in the article. First, railway technicians’ work is characterized by balancing demands for train safety, punctuality, and production against their own safety (De la Garza and Weill-Fassina, 1995). This balancing is to a large extent accomplished through their own efforts and understanding of their work as part of a complex and highly interdependent socio-technical system. Second, accidents and incidents represent breakdowns of understandings of risk and the corresponding means of addressing them. Incidents are breakdowns that do not involve damage or injury (Turner and Pidgeon, 1997, Suchman, 1961). Third, incident-reporting schemes are a kind of organizational communication, and they can be used in similar ways to storytelling to promote organizational learning and prevent accidents (Coan, 2004, Weick, 1995).

Before proceeding with my analysis, three major issues have to be addressed. First, occupational health and safety incidents in railway maintenance are less frequently reported than those related to train safety because they are regarded as less important. Analyzing why this is the case may provide a partial answer to the problem of the low rate of reporting. Second, the absence of injury makes it possible to hide an incident. But why is it considered important to be able to hide an incident? What is worth hiding? Instead of comparing the reasons for reporting traffic related incidents or occupational health and safety incidents, I will focus on analyzing occupationally-based rationales for reporting or not reporting occupational health and safety incidents to an incident-reporting scheme, without regard to the reporting of train safety incidents, but instead with regard to corporate practice regarding incident reporting. Third, all the empirical data drawn on for this article dates from 2000 to 2003, the years before or just at the beginning of the period when Synergi was introduced. However, because no means have yet been put in place to address the obstacles to its use that are identified in this article, the conclusions drawn here are probably still valid when it comes to explaining the continued low frequency of reported occupational health and safety incidents after 2003.

As in the case of corporate responses to accidents, employee understandings and repair of accidents must be accounted for not only in terms of the accidents themselves but also in terms of the cultural frames and material practices that preceded the accident, and were disrupted by it. In order to be able to analyze occupationally-based rationales on their own terms, rather than as irrational anomalies, there is a need for ethnographic data on the concerned occupational communities and their work. Ethnographic data will throw light on sense making in these communities.

Section 2 of this article is followed by an outline of the underlying conceptual assumptions behind incident-reporting schemes in terms of accident etiology and organizational learning mechanisms. These assumptions are then contrasted with occupationally-based storytelling and accident etiology. Thereafter, the occupationally-based scheme for reporting and analyzing incidents is analyzed as well as the corresponding repair mechanisms using three incident stories collected during my fieldwork. Finally, I present the implications for understanding incident reporting and outline the actions required to learn from incidents in order to prevent accidents.