Near-misses: Were you lucky or were you good?

We think that we can learn from near-misses because we believe that we would recognise the circumstances prior to the event and do something about them in the future. The general definition of a near-miss is that something bad would have happened had it not been for some form of chance or lucky intervention. Notwithstanding this general definition, the problem is that there are two sorts of near-misses: the first is where something could happen, and the second is where something almost happened. Whilst these might appear to be similar, research has shown the way we address them is likely to be cognitively different.

‘Could happen’ refers to a background risk which doesn’t change, no matter how we think about it e.g., the stated risk of a fatality in diving is 1:200 000 dives although the risk for CCR diving is much higher.

The other, ‘almost happened’, refers to the perception of risk and the cognitive biases that influence our perception of it happening to me as a diver e.g., outcome bias and severity bias. If we know someone who has died on a dive, then our perceived risk is different. If we know someone who has run out of gas on a dive due to task loading, we are likely to change how we plan and monitor our dives. At some point, we might even perceive that the risk involved in diving is too high, and we give it up. While the background risk hasn’t changed, the perceived risk in the second one (‘almost happened’) has changed due to cognitive biases.

Near misses (‘could happen’) are often treated as successes, and this can lead us to a normalisation of the risks that we face. Consider the loss of the space shuttle Columbia. The foam blocks from the external fuel tank had been striking the shuttle’s protective foam panels on at least 30 missions prior to the loss of Columbia on February 1, 2003. However, rather than increasing the urgency to resolve the problem, it has been suggested that it led to a level of organisational complacency because the loss of the orbiter was something that ‘could happen’ rather than ‘almost happened’.

If we look at the following dive scenarios, consider if there is a near-miss, how the near-miss is perceived (‘could happen’ vs ‘almost happened’) and whether those involved were lucky or good.

Wreck penetration

Scenario 1: Two divers undertake a swim through of a wreck which involves some simple navigation. There is some percolation, silt, and rusticles. They swim through for about 100m and exit without an issue. They are happy with the result, especially as they got some nice photos.

Scenario 2: Two divers undertake a swim through of a wreck which involves some simple navigation. There is some percolation, silt, and rusticles. They swim through for about 100m and as they exit and the lead diver turns to get an ‘exit’ photo, they look back and notice a fair amount of silt following the second diver out.

Scenario 3: Two divers undertake a swim through of a wreck which involves some simple navigation. There is some percolation, silt, and rusticles. They swim through for about 75m, at this point, the second diver signals that they have an issue, and the first diver turns to help. They notice that there is a large amount of silt because the second diver’s trim and buoyancy have kicked up the silt. They quickly referenced the situation and guided the second diver to the exit point before they became lost inside the wreck. They had a debrief to talk about the situation and how it developed. The second diver said that they knew they had poor propulsion skills, but this was the first time they’d encountered a bad outcome because, in open water, it wasn’t an issue. Besides, they were nearly always last and there was no one behind them.

Instructional Dive

Scenario 1: A new instructor has four students on dive 4 in an Open Water-level class. They also have an additional non-class diver with them who is finishing off some experience dives for a Drysuit class – the dive centre manager told the new instructor that this was okay. The novice divers completed all the drills in the time allocated for the dive – the dive durations were short to meet the shop schedule. The students have demonstrated ‘mastery’ of the skills and are now certified as ‘autonomous divers’. This means they can dive with any other similarly certified diver (in similar conditions) to 18m. The Drysuit diver just followed along and met their objective. They got certified in the class too. The dive centre manager was pleased with the new instructor as they were going above and beyond to meet the dive centre's objectives.

Scenario 2: A new instructor has four students on dive 4 in an Open Water-level class. They also have an additional non-class diver with them who is finishing off some experience dives for a Drysuit class – the dive centre manager told the new instructor that this was okay. The novice divers completed all the drills in the time allocated for the dive – the dive durations were short to meet the shop schedule. The Drysuit diver ended their dive with only 20 bar of gas because they were struggling with the Drysuit. The instructor did not notice this during the dive or afterwards. The Drysuit diver told the OW divers in the pub later that they knew things weren’t good but they did not want to interrupt the instructor as they knew the primary reason for the dive was their OW-level class. The students have demonstrated ‘mastery’ of the skills and are now certified as ‘autonomous divers’. This means they can dive with any other similarly certified diver (in similar conditions) to 18m. The dive centre manager was pleased with the new instructor as they were going above and beyond to meet the dive centre's objectives.

Scenario 3: A new instructor has four students on dive 4 in an Open Water-level class. They also have an additional non-class diver with them who is finishing off some experience dives for a Drysuit class – the dive centre manager told the new instructor that this was okay. The novice divers completed all the drills in the time allocated for the dive – the dive durations were short to meet the shop schedule. The Drysuit diver ended their dive with only 20 bar of gas because they were struggling with the Drysuit. After the dive, during the debrief, the instructor asked each of the students about what went well and why, and what they need to improve on and how they are going to do it. When they got to the Drysuit diver, they were surprised that they had so little gas. The instructor realised that he wasn't paying attention to the Drysuit diver and missed their struggling situation. This was because of the additional task loading that the manager had created. While nothing bad had happened, it could have been worse. The instructor emphasised to the Drysuit diver that while the course was important, their safety was a higher priority. They also spoke with the manager and explained the situation, highlighting that while nothing happened, it almost did and as such, the centre needed to change its procedures.

Why were the divers lucky or good? How much control did they have over the situation, the environment, and the skills/knowledge of the other divers involved? How much were they able to actively manage the situation?

The research indicates that if we have a ‘successful’ outcome following a near-miss, then our perception of the risk is lowered, and therefore we are more likely to erode the safety margins that we have developed through training, equipment configurations, standards, and ‘rules’. This has been likened to a ‘Normalisation of Deviance’ – note that a Normalisation of Deviance isn’t just about the deviance, but also the social acceptance of the deviance (or reduction in risk margins). How many similar scenarios can you think about in your diving or that of your peers?

• Ending a dive with no reserve to be able to do a gas-sharing ascent.
• Doing a visual jump in a cave because you know the cave.
• Being really fatigued on a dive and missing some critical hand signals during the dive.
• Missing the gas analysis on a cylinder because of high workload and missing that it could ave been over the MOD of the gas had it been dived - fortunately the dive was scrubbed.

What can we do about this?

Understanding divers' reactions to near misses is merely the first step towards reducing how these events reduce learning from near misses. The research shows that near-misses are assessed as being more favourable than failures, they lead to minimised perceived risk and can lead to increased comfort with risky choices. There are a number of things we can do to mitigate these issues:

• induce counterfactual thinking,
• raise understanding of how risk perceptions can shift, and
• making probabilities more visible.

Divers with counterfactual mindsets utilise more analytical decision-making processes and can be motivated to learn from past events. They may learn to change their behaviour if they consciously recognise that they were very close to a failed outcome and that they could have acted to avoid it (i.e., not blame circumstances). However, not all near-misses elicit counterfactual ideas and not all counterfactual thoughts result in effective learning. Consider asking simple questions like "Was this dive a complete success? If not, why not?" or "What factors, if changed, would have resulted in a failure, and how robust are these factors to change?". These questions are core to an effective debrief. Although it is really hard not to, we shouldn’t take things for granted when they go right, rather we need to understand why it went right in the manner it did and the best way to do that is via a debrief or some form of reflection.

Reflecting on what went well and why, and what needs to be improved and how, with a specific focus on the issues being covered is therefore critical if we are to learn from a ‘successful’ near-miss. We can ask the simple question, “Were we lucky, or were we good?” If we were good, we should be able to identify what was good and how we can replicate it in future dives. If we can’t identify the factors that led to a good outcome, how do we know we weren’t just lucky? 

Regarding improving understanding of shifts in risk perception and the baseline figures are ongoing projects - they involve the telling of context-rich stories, so that others can understand and recognise the factors leading to adverse events in their own diving, and not just the 'could happen' diving that is out there.

Reference: How Near-Misses Influence Decision Making Under Risk: A Missed Opportunity for Learning. Dillon & Tinsley, 2008

Gareth Lock is the owner of The Human Diver, a niche company focused on educating and developing divers, instructors and related teams to be high-performing. If you'd like to deepen your diving experience, consider taking the online introduction course which will change your attitude towards diving because safety is your perception, visit the website.