Does being more easily seen increase the odds of crashing?

[SVS]

Interesting and counter-intuitive research finding:

"This study was designed to assess the effect of conspicuity aid use on the risk of crash for commuter and utility cyclists. A slightly greater proportion of cases than controls reported using conspicuity aids. There was therefore a raised odds ratio of collision crash involvement for those using conspicuity aids even after adjustment for a large number of important confounders. The study results do not demonstrate a protective effect as expected given previous work testing the effects of such aids on drivers’ awareness of cyclists and pedestrians. This study demonstrates the importance of understanding why many cyclists remain at risk of collision crash resulting in injury despite the use of conspicuity aids."

Here is the link to the research.

[upsidedownmark]

No, it doesn't. Nice attempt at a clickbait title (doh!)

1) Even they don't suggest the high vis is CAUSAL (as title), just that there is a (small) CORRELATION

2) They didn't take a bunch of similar cyclists and dress some in high vis and some not then wait and see - which would be an unbiassed sample, and a 'good' study, instead they looked at crash rates in ppl that did wear high vis and ppl that did not; The only thing that they could have proved is that ppl who choose high vis are more likely to crash (although they note it is statistically insignificant as a trend). That makes sense on many levels - same as 'twunts who wear cameras get into more arguments'. They choose to wear high vis because they're less confident / have got into previous situations (due whatever reason, that suggests they're likely to again), and so on. Vis is not causal, it's a symptom of the person who is (it seems) slightly more prone to having accidents.

[SteveSteveson]

I suspect that there is an issue with who wears high viz. I would like to see that bias removed. I would guess those more likely to wear it are people with less confidence, so more likely to have an accident and at the other extrem people doing higher risk stuff like riding at night or in heavy traffic. A confident cyclists riding on empty roads on a clear day is the least likely to have an accident.

[SVS]

Good point. And another explanation could be risk compensation:

"If cyclists using conspicuity aids are confident of being seen, they may be engaged in compensatory behaviour changes, e.g. cycling in more dangerous circumstances."

I took the quote from this study, which has some interesting stuff about more crashes among people wearing fluorescent clothing, compared to people wearing white or coloured clothing. It concludes:

"Our study is one of very few examining the effect of cyclist conspicuity on incident crashes, but the design did not allow us to account for behaviours of involved parties and road and traffic conditions before the crash. Some case–control studies attempted to address this issue by measuring cyclists’ acute behaviour including use of conspicuity aids before a crash. A Canadian study observed that the risk of collisions with a motor vehicle was increased by wearing fluorescent clothing but decreased by wearing white or coloured clothing. Likewise, a UK study reported an increased risk of collision or evasion crashes by using any item of fluorescent or reflective material. Additionally, a recent experiment in the UK reported little effect of fluorescent clothing on drivers’ overtaking proximities.

Overall, evidence for the effectiveness of conspicuity aids in reducing bicycle crash risk remains equivocal. Some have argued that cyclists’ misconceptions about their conspicuity and subsequent risk compensation could play a role in minimising potential benefits. In an Australian study, cyclists overestimated their night-time visibility and occasional cyclists were more likely than frequent cyclists to do so. There were also misjudgements on the conspicuity benefits of fluorescent vs. retroreflective materials at night. If cyclists using conspicuity aids are confident of being seen, they may be engaged in compensatory behaviour changes, e.g. cycling in more dangerous circumstances."