Even though organizations may have a proper incident investigation process in place, often times the process is less than optimal in achieving its objectives. A good incident investigation process identifies root causes and how to address them while also demonstrating seriousness about and concern for safety so that the safety culture is strengthened. Unfortunately, even when sound technical approaches are used there are aspects of the incident investigation process that fall short.
Five of the more common issues are described as follows:
| Unrealistic timeframes |
| Failing to differentiate based on potential |
| Narrow focus |
| Undermining the culture |
| Ineffective communications |
An interesting and confounding aspect of many incident handling processes is the requirement that an investigation be completed in 24 hours. This requirement may have begun due to the importance of collecting as much information as possible within a 24-hour window because things like people’s memories and conditions at the incident location can change. Another reason this rule might exist is belief that it demonstrates management’s seriousness about safety. However we need to consider the unintended consequences of setting up a process suggesting that a high quality investigation is desired, but is limited to that brief time frame.
The limitations of this approach can be seen in the following example:
An employee reports being sprayed in the eye when a line she was breaking released a chemical. The employee reports that she was not wearing goggles, because when she went to get a replacement for the scratched and oil-smeared ones, the storeroom said that they didn’t have any in stock. The supervisor responsible for doing the incident investigation goes to the storeroom supervisor who says there was a change in the inventory policy, and that the policy came from the accounting department. The supervisor goes to the accounting manager who says that the inventory policy came from corporate to help control cash flow. In addition to the policy issue, the supervisor must look into why equipment is being left in such poor condition and consider how to fix that system. Finally, the supervisor must consider the cultural issue implicit in an employee encountering barriers to safe performance but continuing with the job.
Given the complexity of even this simple example, completing an investigation in 24 hours, including development of quality action items and methods for assuring that the action items have been effective, seems unrealistic. The good intentions that mandated quick investigation can have the effect of ensuring that root causes are not identified or addressed.
Failing to differentiate based on potential
It is not practical or necessary to do an in-depth incident review in every case, but it is important to do so when there is the potential for serious outcomes. A good incident handling process assures that the appropriate level of review is given to every incident by following several important principles:
1. The expectation is set that all incidents are to be reported, but clarify that all reports do not necessarily mean there will be a detailed root cause investigation.
2. A mechanism is created to evaluate whether the incident had serious injury or fatality potential.
3. For low-potential incidents, an abbreviated investigation is used to assure there are no immediate items that need to be addressed to prevent recurrence, or collect data and on a routine basis look for a high number of low-potential incidents that have similar causes.
4. For high-potential incidents, investigate each one, but apply more rigor to the investigation and development of the action plan. If another high-potential incident happens, determine if the incident(s) involved the same root causal factors. Review the status of the action plan and see if it is adequate or if it was fully implemented.
In most organizations each incident investigation is conducted, reviewed, and communicated as an isolated event. When that happens, the opportunity is missed to conduct longitudinal analysis – looking across multiple incidents and incident investigations to identify common themes and common factors. Failing to look longitudinally can result in addressing one manifestation of an underlying issue without addressing the underlying issue itself.
In the example used earlier, the investigation identified a corporate policy that had gotten in the way of safety. If we looked at a group of investigations and saw several examples of various corporate policies emanating from different corporate departments that each in its own way impeded safety, we might recognize a need to systematically address the process used for policy setting.
When an incident occurs, it is important to learn from the individuals involved what happened and why, so that the causes can be understood and addressed. However, the process of being interviewed following an incident can sometimes feel to the injured party like an exercise in placing blame. When information gathering causes the injured party to perceive that the objective is only to find something that person did wrong, that person and his colleagues will perceive that the organization is not truly interested in making work safer, or that the organization values them. When this happens it becomes more difficult to get good information, and workers will stop making discretionary efforts to improve safety and other aspects of organizational performance.
Information should be gathered in a collaborative way. People responsible for incident investigation must be open minded in seeking facts, and not allow their preconceptions or biases to influence what they ask or how they ask questions.
Reports on the results of investigations of serious incidents are likely to be complex and involve extensive information, from interviews to data analyses. It is easy for people to get lost in the minutiae and lose sight of what is truly valuable information. While documenting the investigation is important, when communicating the results it is important to focus on the big picture about the root causes and how to address them. This places attention where it belongs—on how to prevent recurrence of this and similar incidents.
Conclusion
Incident investigation should occur in a way that reinforces the organization’s desire to take care of employees and continuously improve safety by reducing exposure. Creating an incident-investigation process that achieves those objectives requires pre-planning and the avoidance of pitfalls. No matter the process in place, incident investigations should all have one goal ― the safety of the employee.
This post is an excerpt taken from BST’s newest book, The Manager’s Guide to Workplace Safety. This book is designed for everyone who manages people, from the senior executive to the first-line supervisor. Understanding and using the guidance in this book can help every manager to be more effective in driving safety excellence. For more information and to order the book, please visit www.managersguidetoworkplacesafety.com.
