Conventional wisdom suggests investigating and circulating knowledge of near misses. These ‘free lessons’, so it goes, are supposed to help us learn without the need for injury or damage.
But can near misses also lead to a desensitisation of risk over time, focusing on the achieved success, rather than the near loss?
Today’s paper is from Dillon, R. L., & Tinsley, C. H. (2016). Near-miss events, risk messages, and decision making. Environment Systems and Decisions, 36(1), 34-44.
Feel free to shout me a coffee to support my site & podcasts: https://buymeacoffee.com/benhutchinson
Make sure to subscribe to Safe As on Spotify/Apple, and if you find it useful then please help share the news, and leave a rating and review on your podcast app:
Spotify: https://open.spotify.com/episode/7kxAQAnlq5Bg4jKm0GAtTz?si=blBdBWb9QaW0whHrYeK3_Q
I also have a Safe As LinkedIn group if you want to stay up to date on releases: https://www.linkedin.com/groups/14717868/
Transcript:
Common logic and decades of safety practice tell us that reporting and investigating near misses makes us safer. It makes sense, right? If we learn from the almost disasters, we prevent the real ones. But what if that common wisdom, when applied in certain ways, actually creates a dangerous blind spot?
Hello everyone, I’m Ben Hutchinson and this is Safe As, a podcast dedicated to thrifty analysis of safety, risk, and performance. Visit safetyinsights.org for more research.
Today we’re exploring a 2016 paper from Dylan and Tinsley called “Near Miss Events – Risk Messages and Decision Making” published in Environment Systems and Decisions. They studied how near miss events and their reporting might actually increase our exposure to risk rather than always improving learning and safety.
What were their methods? Well, it used two experiments with participants. Experiment one was the cruise dilemma. During the first experiment, participants read a scenario about a planned cruise potentially being impacted by a hurricane. They were randomly put into one of three groups:
- The control group: Told that the cruisers can be diverted by bad weather, but they weren’t given any info about past hurricanes in the area. Simple and neutral.
- The resilient near miss group: Here, participants were told about hurricane forecasts, but crucially, their last three trips were not diverted. This is where people perceived that disaster was successfully avoided, often leading them to discount future warnings.
- The vulnerable near miss group: A similar scenario, but this time, a close friend had their cruise diverted due to a storm. This scenario is where people perceive a disaster came very close, and it encourages them to seek out future warnings and take mitigating actions.
So what did they find?
In experiment one, when participants had information about a prior near miss, like a cruise almost being diverted, but they had no specific information about problems that almost occurred (the resilient near miss group), they overwhelmingly chose to take the risky option to continue the cruise as planned. This was compared to both the control group and those in the vulnerable near miss condition.
Interestingly, thinking the likelihood of problems was lower did lead people to choose the cruise, but the type of near miss information didn’t actually change their perception of how likely a problem was; they just felt differently about the risk. Also, risk-seeking personalities were, predictably, more likely to choose riskier options, but this trait didn’t interact with the near miss information itself.
Crucially, the type of near miss info did impact general optimism. Changes with resilient near miss info – seeing averted disaster as a positive, “we handled it” – made participants more optimistic and more likely to go on the cruise.
What did they find in experiment two?
This expanded on the resilient near miss data, looking at how framing of uncertainty or danger influences decisions. They found that when near miss information highlighted the likelihood of a negative event or its potential consequences, people were more likely to plan mitigation actions, especially compared to receiving no specific risk messaging. This suggests that the messaging matters.
So what do we make of the findings?
These experiments strongly suggest that near miss events and how we report them may contribute to a state of heightened risk (the near failures), or inadvertently, a dull perception of risk (the near successes), or how we narrowly created success. Perhaps our decisions can slowly drift towards the margins of safety over time. Why? Because some near misses actually dampen our feelings of vulnerability to harm rather than sensitizing us to them.
The authors’ suggestion for emergency managers and safety leaders to overcome this negative impact is that when we communicate about a near miss, we should highlight how the event was almost a bad event, emphasizing it as a near failure rather than a near success. People with prior near miss info that highlights the averted disaster as a success are less likely to opt out of a potentially hazardous trip (the near success). This isn’t necessarily because they update their logical estimates of the likelihood – there was no evidence of this – but rather they feel differently about the statistical risk, becoming more optimistic and less cautious.
Ultimately, these findings urge us to rethink how we learn from near misses, ensuring they truly sharpen our awareness of risk, not dull it. In fact, these authors found these results to be fairly stable across other organizations and roles also, so they have other research that confirms this. So this doesn’t mean we shouldn’t report, discuss, or act on near miss events, but should consider how we frame them, particularly within the sphere of safety-critical domains. There’s also a lot more stuff to learn from in daily work that doesn’t rely on rare events, but that likely has its own caveats too.
Limitations: Really one of the core limitations is that they only looked at intended behavior rather than actual behavior, but in saying that, intended behavior has been shown in other work to be a good predictor of actual behavior.
In summary, according to this research, we should emphasize the near failure rather than the near success.
SafetyInsights.org 