Fact-Finding Techniques for Incident Investigations

If you’ve been a safety professional or an operational manager for any significant amount of time, you’ve probably had your share of safety-related incidents. The most significant incidents are usually measured by their consequences. These may result in death, serious injuries, lost or restricted workday cases, OSHA recordable cases, first aid treatment, and/or equipment or property damage. Other incidents are commonly referred to as near misses, where serious consequences like the ones previously listed could potentially have occurred, but, through luck or circumstance, did not. Regardless of the type of incident, there is always one question that is asked afterward: Why did this happen?

People start looking for the answer to that question shortly after the initial recovery actions from an incident, such as caring for the injured, putting out the fire or cleaning up the spill. Most businesses call this quest for information their root cause analysis process. Every process likely contains multiple steps, and a good process will be tailored to fit the business in which it is being used. Most incident investigation processes contain the following basic elements:
• Notification. You can’t investigate and correct what you don’t know needs to be fixed. Most incidents are self-disclosing, like injuries, fires and equipment damage. Some incidents, like near misses, require employees and management to recognize the potential for harm or damage and report it so that it can be investigated.
• Collecting information. All of the information about an incident or near miss is hardly ever immediately available. In mature safety cultures and systems, it is rarely one safety barrier that fails and results in an incident, which makes it necessary to collect more information. Also, most incidents involving human performance are not just the result of someone making a mistake. The information gathering process must be comprehensive to ensure that all of the potential causal factors and conditions under which the incident occurred are considered.
• Analysis. In this part of the safety investigation process, the gathered information is organized into a logical sequence and causes can be objectively determined. Many large companies utilize one or more of a number of commercially available products (e.g., TapRooT, REASON, PROACT) for consistency and trending of identified causes.
• Correction. In a safety investigation process, the ultimate outcome is to identify gaps (causes) in operational systems and make changes to prevent the incident from recurring, or to prevent the same latent causes from resulting in another, different incident. Continuous improvement is the goal of every business or operation, and improvement initiatives must be based on reasonable actions.

Note that a safety investigation is very different from other types of investigations – such as criminal or civil – where human performance is involved. A safety investigation is based on a simple premise that good people were trying to do the right things, but a bad thing occurred. In other words, a safety investigator must start with the belief that no one came to work the day of the incident with the intention of causing an incident. Clearly, if it is determined that an employee did come to work with the intention of causing damage or harm, a criminal investigation is warranted. These types of investigations are not based as much on correcting causes, but more on assigning fault and determining accountability.

Potential Sources of Information
While the details of individual safety incident investigation processes may differ slightly, all of them are intended to be based on complete and factual information, and the information gathered during a safety investigation process is the cornerstone on which everything else is based. Without unambiguous, unbiased information, it is highly likely that all or some causes of the incident will go uncorrected and the incident – or some variation of it – will be experienced again. To quote an old axiom, inserting garbage into the process will result in receiving garbage out of the process.

Collecting information may seem simple enough, but it is not as easy as some think. A safety investigator needs to be able to identify all of the available sources of information about an incident and then extract that information from the identified sources. Listed below are some – but not all – potential sources of information and how they may benefit a safety investigator in determining all causes of an incident.

Physical Evidence at or from the Scene
There is no way to measure how important it is for an investigator to visit the place where the incident occurred. To truly understand what happened, a safety investigator needs to gain a firsthand perspective and appreciation of the environmental conditions in which the incident took place. All too often, a later investigator will see a critical detail or identify a critical relationship between placement of personnel or equipment that had been missed by others. Physical evidence also includes, but is not limited to, the protective equipment an individual may have been wearing, tools that were being used, and the equipment on or with which the personnel involved in the incident were working.

Diagrams, Sketches, Prints and Plans
A cause for the incident can sometimes be discovered in how the system or equipment involved was originally designed or in any design modification prints. In other cases, an investigator will need to see if as-built drawings or prints being used by the personnel accurately depicted the as-found conditions of the work area.

Phone/Radio Recordings
Many utility companies automatically record phone and radio traffic to and from their call centers, dispatch centers and system operating control rooms. In addition, emergency call (911) traffic is also recorded. Obtaining these recordings could help a safety investigator determine the causes behind a misunderstood verbal communication, a lack of communication, the absence or presence of repeat-backs (if required), and/or if standard terminology was used in the time leading up to, during or after the incident.

Photographs/Videos of the Incident Area
In the moments immediately after an incident, everyone should first be concerned for the care and well-being of those affected. It is absolutely imperative, however, that someone make a photographic record of the incident as soon as practical. Invariably, something will get moved or altered in some way if too much time goes by or if the scene is not or cannot be adequately preserved. Also, photographs after the incident are just one part of the total amount of information available. Photographs and video before or during the incident may be available as well. Available sources provide different estimates, but an average resident in a city is likely to be photographed or filmed anywhere from 30 to more than 100 times per day, and that’s not counting videos and pictures taken by pedestrians or passersby with cell phones and tablets.

Records
If an equipment failure or malfunction was part of the incident, an investigator must obtain the purchase records, manufacturer’s instructions and maintenance records of that equipment. These records may contain valuable information that could help determine a predictive failure that wasn’t identified, or show that vital preventive maintenance on the equipment was not recommended, required or performed. Maintenance records could also include X-rays and thermal imaging of the equipment.

If an apparent lack of understanding of safe operating procedures by a person is identified, the investigator must look at the training records for both that employee and other employees to determine if the training was effective or if records of understanding (tests) or proficiency (practical exercises) may be an issue.

Finally, medical records of the employees involved may need to be examined. Some companies mandate post-incident drug and alcohol screening, which can provide valuable information. Other than the involved employee possibly being under the influence of drugs or alcohol, a medical records review may show that the employee could have been taking a prescribed medication or had a medical condition that could help explain why the incident occurred.

Procedures or Policies
It is no surprise to anyone that corporate policies and procedures are not perfect and can’t possibly cover every situation in which employees may find themselves. But if there is a policy, the investigator needs to know what it says, what the intent of the policy is, and if the policy or procedure was adequately disseminated to the workers involved in the incident. In some cases, an investigator could find that there is no policy and that one needs to be developed.

Logbooks, Inspection Sheets, Pre-Job Briefings and Job Safety Analyses
Logbooks and inspection sheets pretty much speak for themselves, but a pre-job safety briefing or job safety analysis can give an investigator keen insights into the thoughts of the employees who were involved in the incident before it occurred. In some cases, the investigator may find that a hazard was not recognized or that it was known, but was not adequately mitigated.

Social Media Information
This is a relatively new source of incident information, but there are documented cases in which social media sources have produced valuable information about an incident. Admittedly, most of the ones that have made newspaper headlines relate to criminals boasting about their crimes, but it is possible that someone may have uploaded a video, picture or narrative about the event that could be helpful.

People
Most safety investigators acknowledge that people are usually their best source of information, but also their most unreliable source of information. Data about an incident can be obtained from involved personnel, victims, witnesses, emergency care professionals, police, former employees and subject matter experts.

The reason why people are the best source of information is because they were actually there to witness the incident. They saw it, felt it, smelled it, tasted it and heard it happen. The reason why they are also the most unreliable source of information is because, well, they are people. The investigator who is trying to obtain information from these people is a person, too. People have emotions, biases, conflicts and anxieties, especially after an incident. Some people will volunteer to take all of the blame for an incident and others will try to blame the incident on everyone and everything but themselves. Some people downplay the incident to help them better cope with it and others exaggerate the incident to draw more attention to themselves. None of these scenarios is helpful to a safety investigator. We need facts, not blame or deflection.

Conducting an Effective Interview
When it comes to interviewing people, the investigator’s challenge is to extract factual, unbiased and dispassionate information from a biased, passionate and emotional creature. How do we do that?

The first thing for a safety investigator to do is acknowledge that all humans have emotions, biases and insecurities. That may sound easy to do, but it isn’t, especially with a utility worker. The prototypical utility worker comes off as supremely confident, fiercely independent and unaffected by human emotion. This is not necessarily a good thing when a safety investigator is trying to examine a human performance issue.

To start the information gathering process and to prepare to interview a person, a safety investigator should do everything they can to make it seem less like an interview and more like a discussion. Here are some tips and pointers for interviewing people:
• Be prepared. Don’t fly in and start asking questions without a plan, but don’t wait too long to get started either. As time passes, personnel involved in an incident or witnesses can:
  o Forget. We all forget things, and sometimes we forget the details first and only remember the big things that happened. Details are vital.
  o Get it wrong. For example, how big was that fish you caught when you were a kid? Over time, people can either minimize or exaggerate facts or get things out of sequence. The order in which events occurred during an incident is extremely important.
  o Rationalize or justify. People may downplay or explain away critical details of an event because what seemed to be significant at the time starts to seem more inconsequential as time passes. Therefore, they think it’s too trivial to mention.
  o Become influenced. No one is necessarily strong-arming a person to lie, but outside influences, ulterior motives, and other agendas may enter into the person’s thought processes, and they might omit or change information about the event based on how it might make them look to others.
  o Become conflicted. An example of this is when someone starts thinking things through and determines that a potential outcome of the incident investigation could get someone else in trouble. This is highly dependent on the safety culture of the company for an internal investigation, and more likely during an external investigation.
• Make a list of key questions. Make sure your questions focus on who, what, when, where and how. Try hard not to ask a why-related question. The “why” part of the investigation is the root cause analysis, and at this stage of the process, it is too early for that. It’s also unfair to ask someone a why-related question when they may not have all the facts and you will only be getting their opinion, not an objectively determined conclusion.
• Start with an introduction. Tell the person up front what you are doing, how you are going to do it and what the ultimate goal of the safety investigation is. Never assume that they know. They need to be a part of the process, not the subject of the process. If you make them feel like they are going to be a key component in preventing the incident from ever happening again, they’ll be more likely to engage completely and share valuable information.
• Speak with dignity and respect. The person you are interviewing may have just gone through a traumatic experience. They are also your key to success. Investigators must honor that and understand that witness contributions are valuable.
• Close with a review. Make sure that what you heard is what they said. Go over the information again. It’s amazing how much more detailed information you can get when you do this. If something isn’t clear, the person will volunteer more information until they are comfortable that you know the information in context and from their perspective.

Throughout the interviewing process, emotions may be high. It’s important to note that the witness will not be the only one experiencing significant emotions during this time – the investigator will be, too. Investigators can be intimidated, led astray during the interview, unsure about the processes being examined or the work performed, and uncomfortable in a role that may make it seem like they are criticizing someone who just had an incident. But if everyone involved knows they have a part to play in a problem-solving process and that their role is vitally important, all parties involved can help each other succeed. The investigator must be the calming force during the investigation and approach the process as a problem solver.

Hopefully, you can see how much time, effort and dedication are needed during the information gathering phase of a safety incident investigation. Without good information, the analysis will be either wrong or incomplete, the recurrence prevention measures that will be implemented may not correct the underlying causes and we are doomed to repeating the mistakes of the past. With good information, we can solve problems and achieve our goal of continuous safety improvement for the benefit of our employees, our companies and the communities we serve.

About the Author: Ron Joseph, CUSP, is the operations safety manager for Dayton Power & Light, an AES company. He is a trained incident investigation team leader and a root cause analysis expert who has served on the TapRooT Technical Advisory Board for more than 15 years. In his 30-year career in the nuclear and utility industries, Joseph has personally investigated hundreds of safety incidents and performed safety audits throughout the U.S., Europe and Africa.