Incident Report Effectiveness

The goal of workplace investigations is the identification of effective incident prevention strategies, and the incident report form is the heart of the process. The form should guide you through a series of logical investigation steps, and serve as an efficient data collection tool.

The majority of investigations are not effective in preventing repeat incidents. A number of factors account for this including policy, procedures, training, communications or management support to name a few. A logical place to begin an analysis and in turn find improvement opportunities in your investigation program is right on top of your desk - the incident report.

Let's take a look at your current report form and a sample of at least twenty-five completed incident reports and review them in a critical and objective manner. Consider the following for review:

• Does the report format address the needs of our workplace?
• How well are reports being completed by investigators?
• What's causing or contributing to incidents?
• Are we fixing the problem?

Few organizations have taken the time to develop an incident report format based on the needs of their workplace. Forms are typically borrowed from other organizations and this becomes very obvious when you see references to processes or departments that don't exist. Using these forms almost certainly creates questions that are “not applicable.” Every question on the report should be fully applicable. If investigators know they can use N/A (not applicable) as a response it may become a default action for many questions.

Is there a strong link between the incident report and the requirements of company investigation policy and procedures? Are clear easy to understand directions included on the form or in a companion document. Incident reporting is not a frequently performed task and some coaching on the form would be helpful. Are examples included for items such as causal factors, and most importantly does it assist you in identifying root cause(s)?

If you are going to develop a new report format it's critical to survey the end users to see what will work for them. In one organization an effort to develop a best-practice incident report resulted in a four-page document. A report of this length is not uncommon in large organizations; however, it met strong resistance from supervisors, as in the past they had never been required to provide this amount of documentation. The safety department and supervisors fought for years over the accuracy and timeliness of completing incident reports. In the end, the safety office worked with supervisors to develop a single page incident report. The result being the safety office was inundated with incident reports, and in turn, was able to take action on these to significantly improve the organization's injury prevention initiatives.

Now turn your attention to the completed reports in your sample. A good indication of a supervisor's ability or willingness to complete the report can be judged by the amount white space on it. There's usually a lot! For example, the question about Environmental Conditions may have no answer. Describe how the accident happened has five lines available for the response, and there's five-words, “John fell off the ladder.” While incidents are often described as a single event, they are in fact a combination of events and factors and these need to be uncovered and documented.

What questions are not being answered? Is this happening on reports from just a few departments, or throughout the organization? What types of incidents seem to have the least or most amount of detail? Now do a little investigation of your own. Survey those that have completed reports as to the difficulties they have encountered.

Why worry about a few unanswered questions? Aren't we focusing a lot of attention on the clerical or paperwork part of the investigation? On the contrary, the report is a critical component of the investigation process. Either you have it right and complete, or you don't. You can't arrive at accurate conclusions or develop solid recommendations based on vague or inaccurate information. If the answer to a question appears to be one that's readily available, this may indicate an investigation shortcoming. If the investigator felt the question wasn't applicable, or didn't answer it for some other reason, you need to find out why.

Put yourself in the position of someone totally unfamiliar with the event. Does the report tell the whole story or does the reader have to use his imagination to fill in the blanks. Management usually reviews incident reports and they form judgments about both the practicality of the recommendation, and your ability as an investigator. What opinion are you creating in their mind?

One way of improving report quality is in the development of a scoring template and the provision of feedback. Assign a numerical value to each part of the report with the most amount of marks awarded to those areas with the greatest impact on prevention, likely the area of root cause, conclusions and recommendations. For example, you might allocate ten percent of the marks for completing the who, when and where aspects, and twenty-five percent to the events description and so on.

Developing this reporting scoring template with the safety committee or supervisors makes it a legitimate tool for providing feedback and setting performance standards. Determine an average score based on what you found in your current sample and set some goals for the future. Some organizations have the scoring grid built into the margin of their report form allowing investigators to measure their own level of performance. In time, anything less than 90% may be considered unacceptable.

It's one thing to have the form complete; accuracy also needs to be considered. Every causal factor identified in the report must have supporting evidence documented. If Housekeeping is identified as contributing to the incident, what exactly does this mean? What is the housekeeping standard? What evidence do you have and where did it come from?

As you review the reports keep a list of the contributing factors. You will likely find the Pareto Principle at play, in other words, the 80/20 factor. The majority of incidents will be caused by a small number of factors. The fact that you have multiple reports listing the same contributing factors is in itself an important message. Obviously, prevention efforts to date based on these investigation reports have not been successful.

You can use this as an opportunity to reinvestigate, at least as a paper exercise, all the situations with a similar set of contributing factors. Future incidents with the same contributing factors might call for a broader or team based investigation to better ensure the identification of effective corrective actions.

This final and perhaps most telling factor is that of recommendations. This is where the rubber meets the road. In your sample group you should not be surprised to find the following. First, the majority of recommendations have never seen the light of day. Although they were developed as being the right solution to the problem, they never got off the paper. Secondly, many of those that did make it into the workplace likely enjoyed a very short life. The department has fallen back into their old ways, and incidents are repeating themselves. How many of the recommendations listed are in place? Why not? These two questions will speak volumes about the quality of your investigation program, or may point out an even larger organizational problem.

The single most important thing an investigator should do to ensure their recommendations are effective in preventing future incidents is to test their recommendations before sending their report forward. Not doing this may be one reason that management has not responded to recommendations of past incident reports as the recommendations may have a track record of possessing little value.

One method of testing is to use the risk matrix developed in your hazard assessment process as an evaluation tool. It has risk categories that range from low to high using numbers, letters or some combination thereof. If the incident under investigation was found to be a high-risk activity, your recommendations when in place should now make this a medium or low risk situation. Or, it may remain high-risk, but the frequency of it occurring has been diminished significantly. All recommendations should identify in a measurable manner exactly what is to be done, who is to do it, and when it will be done. A recommendation to “clean up the air in the welding shop” fails by all these measures.

The incident report is something that most of us deal with on a frequent basis, and perhaps take for granted. In doing so we have not recognized it's full potential as an injury prevention tool. Performing some of these simple measurements will put your report in a whole new light. You will likely find that improving a form that is not meetings users needs will meet with very little resistance, and will in fact, be welcomed. At the same time it will help ensure what you have always said investigations were designed to do - prevent future incidents.

Back to Articles