You don’t need to be called Sherlock to investigate your own accidents

What’s the issue?

Over the years we’ve seen dozens of accident reports but very rarely do we see a thorough investigation leading to a satisfying ‘that’s never going to happen again’ conclusion and action plan.

Here is a summary of the problems we see and the reasons behind them:

Stopping too soon

There may be lots of reasons why someone tasked with investigating an accident ‘stops too soon’. For example:

  • time pressure from other pre-existing work
  • pressure from a senior manager
  • moving too quickly to a solution without adequately investigating the causes
  • ‘social pressure’ (within, or outside of, the workplace) e.g. ‘I’ve bothered them enough about this already’ etc.

Another reason for bringing the investigation to an early close may be the realization that accident causation can be complex and requires a high level of analytical skill and so it’s easier to settle on solutions based on past experience. Often when the circumstances leading to the accident are not adequately described at the beginning e.g. on Accident/Incident Forms, in an effort to complete the job assumptions are made to fill the information gaps.

Managers will sometimes assume that the ‘cause-leading-to-effect’ relationships are obvious (‘We all know what’s going on – now how do we fix this once and for all’). Whereas those nearer the problem may silently be thinking about a whole range of different causes e.g. poor communications; cheap parts; the operator or contractor was to blame; the procedures were impossible to follow; too much pressure from Management, etc.

The ‘Blame Game’

In the workplace, we sometimes place blame because we don’t know what else to do when dealing with the aftermath of an accident. Most supervisors and managers are not adequately prepared for this aspect of their job and so assigning blame can be the quickest and easiest thing to do.

Blame allocates both the fault and the responsibility to the individual and doesn’t acknowledge faults inherent in systems and the responsibility of those who set them up.

For the conscientious worker, unjustly assigning blame is a form of punishment. Indeed, the belief that punishment will improve long-term behaviour in adults is not supported by any facts or studies. Disciplining or firing an employee blamed for an accident may be the worst thing to do since if it is unjust they will bear a grudge and they may well have good insight as to the real causes and how to prevent a recurrence (and may well now keep this to themselves).

It’s true that when we commit an error with purpose, knowing full well it violates established rules, we can accept the consequence of our actions even if this includes being disciplined. However, this kind of deliberate error is very rare in the absence of other endemic problems (e.g. problems with the workplace ‘culture’).

Much more likely is an error is one of omission e.g. we omitted to do something because of lack of knowledge. In such cases if we are punished, we rarely see the value of the punishment and so will not change our behaviour. This is because we feel like we are unjustly being blamed for our behaviour rather than supported to gain the necessary knowledge to prevent a recurrence.

If human error is believed to be a significant cause, the reasons for this must be investigated. Of course, lack of knowledge, training or unsuitability for the job may be the causes of this error, but these are primarily management and not operator failings.

The ‘one root cause’ myth

A myth has been created that there is always one ‘root cause’, however there are usually many different avenues containing contributory ‘causes’. These contributory causes will often have a variety of people whose actions inadvertently played an important part.

So effort can be wasted by investigators looking for – and debating between themselves – which is the definitive ‘root cause’. Committing to avoiding the use of this terminology also helps to diffuse the ‘who’s to blame’ reflex.

It is better that investigators search widely for all possible solutions, and then select the ones that are within the control of the organisation to implement and will effectively prevent a recurrence.

The illusion of ‘common sense’ and a single reality

Common sense is a term mainly reserved for those who have the benefit of hindsight after an accident, or used whilst in the process of setting someone up to fail e.g. by not providing the right equipment to do a job (‘… just use your common sense…’).

However, the term ‘Common sense’ cannot be defined and is an ‘oxymoron’. Intelligent people who would be said to have ‘common sense’ prior to an accident are dispossessed by those standing in judgement after an accident. They are often guilty of simply trying to deal with unknown situations, follow someone else’s faulty instructions, or just of making perfectly normal human errors.

Needless to say, this populist catch-phrase has no place in accident investigations.

Teamwork can iron out some of our biases

Many studies have also shown that we all have our own unique set of senses, knowledge, and strategies for processing information. So the conclusions we draw from witnessing the same event can be very different from one another.

After many historical miscarriages of justice, the Police have adapted the techniques they use in investigating crimes to take account of this i.e. ‘the truth’ of what happened is more likely to be found by adopting a non-judgemental, enquiring approach, and then searching for commonality between the differing individual accounts recorded.

Alongside our own unique view of the world is our own ingrained collection of biases and prejudices – we tend to subconsciously favour conclusions that fit into our own pre-existing habits or beliefs, sometimes despite direct evidence to the contrary. So carrying out a meaningful accident investigation can be as much about the investigators challenging themselves rather than just making routine set of enquiries.

For all of the above reasons, it’s a mistake to see an accident investigation as being one person’s job: in reality unless people work and cooperate as a team then very little will be achieved. The diversity each person brings to the process provides the opportunity to consider a bigger and clearer picture of each situation.

So what can we do to combat these pitfalls?

 In our consultancy we try to steer anyone investigating accidents away from the above pitfalls – here is a quick summary of the basic concepts:

  • Management should demonstrate care and concern towards employees – during an accident investigation this is by avoiding a ‘blame culture’ in favour of a ‘just culture’ i.e. levelling accountability only after assessment of all of the facts and influences behind actions or omissions at all levels
  • Those investigating the incident should have a good understanding of the mechanism of human error, so that if ‘culpability’ is to be assessed it is done intelligently and in the light of this knowledge
  • Employees should feel that they are able to report issues or concerns without fear that they will be unjustly blamed or disciplined personally as a result
  • Confidentiality should be maintained throughout the course of the investigation