Practitioners’ perspectives on incident investigations

This study surveyed the views of industry practitioners familiar with incident investigation processes, the types of controls that are implemented following investigations and also reviewed submitted incident investigation manuals and procedures.

222 survey respondents were included in the data, across mining, construction, transport/postal/warehousing and other industries.

Before moving into the results, the authors provide a summary of a failure to learn, resulting in repeat incidents time after time. They also discuss the phenomenon of what-you-look-for-is-what-you-find, highlighting that the particular worldview we have on accident causality and how it’s described in procedures influence the types of factors that are found in investigations (e.g. if your worldview or method sees accidents due to human error, you will construct errors as causes).

Note that the interpretation of the results may jump between answers that do consider particular things (e.g. “was” considered) and in cases where particular things were not considered (e.g. “wasn’t considered). I’ve tried to bold these instances so it’s obvious which one is being talked about.

Results

Identification of controls present at the time of incident

Over 57% of all industry survey respondents indicated that the extent of controls being present at the time of incident *was* considered in their incident investigation processes.

Mining had a similar response result to the overall finding, whereas construction was higher at 73% where investigations always considered the present controls.

Transport/postal/warehousing was 64.5% and the ‘other group’ was 60%.

Assessment of the effectiveness of those controls present at the time of the incident

For this question, 57.7% of the respondents said control effectiveness *wasn’t* always considered as part of the current investigation processes. Similarly, 60% of the mining and 57.6% of construction sample (respectively) indicated control effectiveness wasn’t considered in their process.

In contrast, 58% of transport indicated that control effectiveness *was* always considered.

Identification of controls that although absent at the time of the incident may have been an effective prevention control

Data across the whole sample indicated that absent controls *wasn’t* always considered part of current incident analyses, at 52.8%.

In mining (57.3%) and other groups (59%), the consideration of controls being absent at the time of incident *wasn’t* always considered.

In contrast, in construction 55% indicated their investigations “always” identified absent controls that if present, may have been effective to prevent or mitigate the event. This was 65% for transport.

Opportunity to learn

The authors provided a summary of the learning opportunities based on control evaluation (see below). The authors highlight that only 61.7% of incident investigations focus on controls, less than 42% focus on evaluating both present and absent controls and “only 34.7% of incidents assess or make recommendations about enhancing the effectiveness of controls” (p193).

Incident manual evaluation results

This question reported on the evaluation of the submitted incident investigation manuals and procedures. One finding was that the “most frequently used tools by investigators do not routinely require arresting and mitigation controls to be identified as part of the incident investigation process” (p195).

Another interesting finding was that only 14 of the 24 investigation documents submitted to the researchers required “a description of the incident, Human Factors considerations and sequence of events to be undertaken as part of the incident investigation process” (p195). I think the absence of human factors integration, at least in the documentation, is quite telling.

For the analysis of attributes related to incidents, the authors found that only incident cause category was clearly considered amongst the submitted documentation, and consequence component to a lesser degree.

They found very little reference to the role of mitigation controls although prevention controls were more readily recognised. However, they note that initiating events or hazards “was neglected by over 60% (62.5%) of organisations incident investigation documentation” (p195).

Authors: Philippa Dodshon, Maureen E. Hassall, 2017, Safety Science

Study link: http://dx.doi.org/10.1016/j.ssci.2016.12.005

Link to the LinkedIn article: https://www.linkedin.com/pulse/practitioners-perspectives-incident-investigations-ben-hutchinson