Chris516

I don't use pedal reflectors at all. I use Planet Bike Super Flash on my back, and a 6-mode red reflector on the seatpost. I did use them at one point, until reflectors fell off the pedals. Then I just decided to stick with cruiser pedals.

Bacciagalupe

They didn't cherry-pick victims. They wanted to get data on crash factors, so they did a survey of crash victims.

I-Like-To-Bike

Yes they gathered data; insufficient data to make any kind "risk study," or any kind of intelligent analysis of bicycling risk.

The use of only hospital data, without any data about the cycling population as a whole, is a worthless "risk study" database for determining risk for the bicycling public. It is also useless for determining the effectiveness of various safety equipment in reducing cycling risk.

jimc101

Don't believe that; surely it's eaten long before it can melt

I-Like-To-Bike

Ice cream NEVER melts. This was proven by a "study" that gathered all its data from ice cream kept in the freezer.

prathmann

Sure, but there are well-known statistical problems with this kind of case-control study. The conclusions are based on comparing the injury severity of those cyclists who were unlit with those who had lighting and ascribing any differences to the presence of the lights.

But there might be a variety of other factors that differ between the two groups. E.g. it's reasonable that cyclists who choose to use lights may be more safety conscious in other ways compared to those who ride without any lighting - they may well be more likely to ride with traffic instead of against it, look more carefully before crossing intersections, choose safer routes, etc. I don't doubt the general conclusion that using lights is safer than not using them, but I wouldn't trust the results of this study to tell me quantitatively how much safer the presence of lights alone makes a cyclist.

SteamDonkey74

Oregon House Bill 63 (1901) sought to require lights for all cyclists, and to establish punishments of $5 fine and/or three day imprisonment for failure to use them.

Well, that law did not pass 111 years ago, but I ride with lights at night and during periods of poor visibility anyway. I see it as a reasonable measure I can take to keep from getting hit, which is no fun.

surgeonstone

The man deserves a Nobel prize who wrote that article.

alhedges

Nope.

I know it is comforting to believe that you can control everything by being prepared, but if an 85 y.o. dozes off while driving in broad daylight, or is even somewhat inattentive, even the brightest light is not going to help you. Cars with three bright taillights, one at eye level, get rear-ended *all the time*.

Obviously cars without taillights are in greater danger, and just as obviously you want to do what you can to be safe. But taillights are not a magical force shield, and even the most expensive setup won't prevent you from being rear-ended.

Bacciagalupe

Did you actually read the study? Maybe you noticed the part where the authors said it's basically a complement to three other databases which cover similar data? Or where it explicitly says that the study "should also act as a guide to and instigator of future related research" ? Or the various qualifiers in the study, which explicitly state that certain specific results can't be generalized to the general cycling population?

Yeah, I didn't think so.

I would agree that the study is not definitive. But it's not supposed to be definitive. Rather, it gives us some useful characterizations of the cyclists who did get into accidents, and opens up further avenues for inquiry. In the interim, it makes numerous useful observations, such as indicating that visibility reduces the severity of crashes; that higher speeds increase the likelihood of a head injury; and that riding in the same area may result in riders letting their guard down, which may increase the chances of getting into an accident at those locations.


They drew a correlation between visibility and less severe injuries. I'd say that is one decent piece of evidence in favor of increasing one's visibility when cycling


Yes, it's a basic issue of conflating correlation with causation. That's why the authors of the study don't posit this as definitive.


...or, we could assert that the riders with lights assume they are safer, and thus ride in a riskier fashion. (Sound familiar? ) That type of conjecture can cut both ways.

Thus you need studies like this, combined with other data, to draw a more definitive conclusion.

FBinNY 

+100,

Being seen isn't a digital phenomenon, as in seen or not seen. There are too many things a play. There's also a big dofference between being visible, and being noticed, as in where a driver registers your existence and adjusts. Big visible things aren't seen all the time, either because of visual or mental distraction.

When it comes to bikes on roads, more visibility is better, but no amount of visibility is proof against being hit.

I'm of the opinion that the various state tail light statutes can work against bicyclists. For example, in New York, red is the only legal tail light color. In some conditions a steady red might appear to be a motorcycle or maybe a car with a single tailight, causing the drivers to misjudge the cyclist's speed and distance.

Years ago I became an outlaw, and adopted a flashing amber as a taillight. I've had police officers stop me and remind me that the lighting wasn't proper, since flashing amber is reserved for stationary road hazards like trenches in the pavement. I politely respond by asking how many times hs the officer seen a car run into a marked trench compared to how many cyclists he's seen hit. Then tell him that given my speed, I'd rather that the motorist behind treat me like a hole in the road and adjust accordingly. Not once has an officer not seen it my way, with most saying they'd rather an "illegally" lit cyclist than an unlit one.

I-Like-To-Bike

"Combined with other data" that is missing from this so-called "guide to and instigator of future related research." Heck, any statement or gathering of anecdotes could be considered a "guide to and instigator of future related research." The data in this so-called "risk study" is insufficient to draw the conclusions about risk reduction presented.

prathmann

Actually the only need for studies like this is to serve as an example of how not to do studies since the data gathered is inherently useless because of the lack of proper controls. So yes, you need other data gathered in a proper study protocol to draw conclusions - but that data should be used on its own and not contaminated by being combined with the data from this study.

limeylew

I find it amazing that anyone would need to do a 'STUDY' to arrive at this conclusion.

catonec 

it is surprising how may riders I see at night w/ no lights and no helmets. usually on MTB's or SS/FG.

Burton

Looks to me like there's a serious drop in IQ in male roadies after the age of 30. The idea of driving without lights at speeds over 30 km/hr on unlit streets in poor states of repair and with possible obstacles is just ...... stupid.

I-Like-To-Bike

Heck, don't need no stinkin' study, with or without other data, to come to that conclusion.

dougmc

Of course, being that they looked at a self selected group of cyclists, that skews the data considerably.

I'm a male aged 35 to 54. I'm not really a lycra wearer, for the record, but I do fit into this age category. And being that I'm a bit older, more fragile and better insured than I was 20 years ago, I'm a whole lot more likely to choose to go to the doctor for something that's not life threatening than I would have been twenty years ago -- and if I didn't go to the doctor, I'd never be counted in their study. And I doubt I'm unique in this.

And really, to decide that this 35-54 age bracket is particularly prone to crashes or that they're likely to be going over 19 mph (which isn't particularly fast, I might add, even for a non-roadie) requires information that's completely unavailable in a study where you only look at the cyclists who end up at a hospital (such as how many cyclists fit into these categories that don't get hurt at all, or do get hurt but don't go to the hospital.)

There's a reason that people are questioning the methodology used here -- because it's highly questionable, seemingly based on poorly done statistics.

Bacciagalupe

To what? People who agreed to fill out a long survey? Is that well-defined group with a bias?


This wasn't riders going to the doctor for a sore knee. It's cyclists who went to the hospital.


Again, hearkening to the NYC study: 91% of bicycle fatalities were men, and fatality rates were highest for men in the 45-54 age range. While it is very likely that men are a higher percentage of riders, the fatality rates for men are almost 10x greater than that for women.

19mph is actually pretty fast if you're in an urban environment like Melbourne. It's also useful to correlate cyclist speed and head injury rates.

Again, the methodology is fine -- as long as you keep in mind it's not intended to be the end-all and be-all of surveys. Nor can a city like Melbourne give every cyclist a helmet cam, and hire an army to sort through the footage in an effort to tease out patterns. It's only a part of the picture.

dougmc

Yes it is. But in this thread we're talking about people who chose to (or were chosen to) go to a hospital.

If you're dead or unconscious, the decision will be made for you. If you've broken bones beyond a finger or toe, you'll probably be going. (Though I've seen young people choose not to go after a broken arm.) But if you hit your head and are a bit woozy, have deep cuts, etc. -- you'll be making a judgement call. And in my experience, middle aged (and older) people are more likely (than younger people) to be hurt in the same situation, are more aware that being hurt now can have long term repercussions in the future, and are more likely to have medical insurance and good jobs meaning they can actually afford to go to the ER. (In the US this is a large factor. I don't know how it works there.)

The specific claim I was referring to was "Looks to me like there's a serious drop in IQ in male roadies after the age of 30." -- and I'm pointing out a few of the that the many ways that this conclusion is likely invalid given the evidence used to come to the conclusion.

In the US I think the figure is around 7x. But unless you can actually quantify just how much a higher percentage of riders men are (and look at how much they ride and where), you can't really use this to determine that "men's IQs drop more" or anything along those lines. Though at least when you're looking at fatality rates you know there's not a self-selection bias, so that helps make the figures more accurate.

Most cyclists can't sustain a speed that fast on level ground, but most places have hills that allow even inexperienced cyclists to go that fast. Certainly, roadies are not the only ones exceeding 19 mph. And considering that the slowest commonly seen speed limits around here are 25 mph -- it's still pretty slow.

And yet that's how such things are usually presented -- as the "end-all-be-all". Not usually by the survey authors themselves, but by those who find that the conclusions (or some specific bullet point) supports their agenda.

Burton

LOL Looks to me like you're pretty good at picking out just what suits your own agenda yourself. That little quote extracted from my entire post is entirely out of context. And thats MY opinion and I'm the one that wrote it.

If you want to quote me - lets at least keep it reasonable. My closing comment was:

And that 'study' indicated that the majority of individuals stupid enough to do that were apparently spandex clad males over the age of 35. If you want to read anything else into it - maybe you should check yourself in for possible head injury. Mmmmmmmm do you wear spandex and drive fast late at night in unlit areas?

dougmc

That "little" quote was 40% of the original content in your post.

And yes, of course it's your opinion. And it was a fine example of an opinion that wasn't really supported by the facts you quoted with it.

What does that even mean?

Um, none of that was in the post I quoted from.

Here is the entirety of your post (excluding the quoted parts) --

As for wearing spandex, I pre-emptively answered that already if you'd bothered to read my post. As for driving fast late at night, I drive at appropriate speeds, day and night. I ride at appropriate speeds as well, day and night -- and if it's night, I'm lit up like a Christmas tree.

I don't know of anybody who drives without lights (at night? You mean at night, right?) I do know some cyclists who ride without lights at night, however, and I think that's dumb. None of them really wear spandex while doing so, however -- the roadies I know that ride at night use appropriate lighting.

Burton

Nope - my post included direct quotes from that study. I just rephrased them on a slightly more entertaining manner.


ALL of that was in the original post. Thats exactly what those quotes from that study indicated. Here's the whole post again.

Quote:

Sixty per cent of crashes did not involve another vehicle, but were caused by a range of factors including hitting tram tracks, potholes or debris on the road, even gusts of wind.

Quote:

that middle-aged men in Lycra - or mamils - were the road's most endangered species on a bicycle. Three-quarters of crash victims surveyed were men and about two-thirds of those men were aged 35 to 54.

Quote:

The study also found that almost half the crashes in which the rider was hospitalised involved a blow to the head, with cyclists who were travelling faster than 30km/h five times more likely to receive a head injury than slower riders.

Looks to me like there's a serious drop in IQ in male roadies after the age of 30. The idea of driving without lights at speeds over 30 km/hr on unlit streets in poor states of repair and with possible obstacles is just ...... stupid.

I'm still thinkin' you should check yourself in ....