UKC

40,000 patients were treated in U.S. emergency departments for rock climbing-related injuries between 1990 and 2007 a new study reports.

Read more at http://www.ukclimbing.com/news/item.php?id=48657

In reply to UKC News:
How can a 2 year old sustain a genuine rock climbing injury?
I suspect the search terms these researchers used to interrogate the database were duff; which then cast doubts on the validity of the rest of their findings....

In reply to UKC News:
"Rock-Climbing Injury Rate Soaring"

I can't see where it gives the injury rates (ie injuries per participant) as opposed to injury numbers. Can someone point me in the right direction?

"The severity of fall-related injuries correlated with the height of the fall."

Who'd have guessed?

In reply to UKC News:

“We found that the climbers who fell from heights higher than 20 feet accounted for 70 percent of the patients there were hospitalized for a rock climbing-related injury,” explained study author Lara McKenzie. “This trend, combined with the fact that rock climbers have a higher hospitalization rate than other sports and recreational injuries, demonstrates the need to increase injury prevention efforts for climbers.”

- erm, no it doesn't. It's a non-sequitur. The first sentence isn't a trend, it's a single data point.

All in all, quite a good demonstration of the need for context when presenting study results.

Y

In reply to Dr Fran (Vagabond MC):
I also suspect there has been some odd manipulation of the numbers. Having the data split in to such wide age groups could be hiding a lot. I imagine there are very few very young people being injured so covering 0-19 years old in one group could be skewing the data.

In reply to Toreador:

Mick links to their press release. Can anyone get access to the original journal paper?

In reply to toad:

I've had a good search on the American Journal of Preventive Medicine site where it is supposedly published, but I can't find it on there. I was quite interested to understand a little more about the nature of the research. From the story summary it doesn't sound like very good research. It sounds like some stats trawling and it would be nice to know the nature of those stats and the manner in which they have been interpreted.

In reply to toad:


I've tried. I can't find it on the American Journal of Preventive Medicine website or via Scopus or PubMed.

I'm very interested in looking at the original data. At first glance, 40 000 incidents over 17 years from a claimed 9 million participants seems incredibly low.

(If personal experience is anything to go by, most of the accidents occurred in Joshua Tree over the labor day long weekend)

In reply to duncan:

Hi Duncan,

it states in the brief article that the study used stats from the

http://www.cpsc.gov/LIBRARY/neiss.html database.

This is the Consumer Product Saftey Council's database called the National Electronic Injury Surveillance System (NEISS). It's supposed to track injuries based upon the products involved e.g ladders.

"every emergency visit involving an injury associated with consumer products"

As this data is all product based rather than activity, it's rather important to know how the researchers have mapped products to activity in order to conclude that the injury is climbing related.

To me it reeks of stats trawling based upon dubious assumptions in order to knock out a quick publication containing spurious conclusions that will grab a headline and some easy PR.

In reply to duncan:

Handing out reprints to interested parties is bread and butter in a lot of forms of science. Mebbe contact the authors direct or the PR people noted on the press release.

Y

In reply to Hugh Cottam:


As does a press release issued before the paper has actual appeared.

In reply to duncan:

Lazy Lara uses the same database to achieve similar conclusions about computers

http://www.ncbi.nlm.nih.gov/pubmed/19524141?dopt=Abstract

In reply to Stig:
>
It's unfortunately increasingly commonplace, and the source looks reasonably credible. Many institutions press release with much less than this. Don't think anyone is to blame outside of the issuing body.

In reply to toad: I know, I work in a similar context and I would just find it embarrassing. But I also think that simply recycling PRs without checking the original source is lazy - but I understand why it happens.

An interesting artical but not very useful. The activity, apparel or equipment for mountain climbing are all lumped together in the database and a very small sample size is used - for 2008 90 injuries are used estimate an activity level of 3386 for the whole of the US. If that is out of 9,000,000 individuals participating during the year you don't have to be too worried about being hospitalised (0.038% chance of being hospitalised). And that's not particually useful either as it doesn't take into account the number of hours people spend climbing every year.

I don't know America very well, but there seem to be very few hospitals contributing to the database on the West side which I thought had the majority of the mountains?

One of the naratives to a mountain climbing injury:
"WENT ROCK CLIMBING, REPORTS USED ?LEAF TO WIPE HIS BUTTOCKS, NOW WITH P RURITIC RASH >> CONTACT DERMATITIS"

Not sure this one should really be counted either!
"SWINGING OFF ROPE AT CREEK SANDBAR PT FELL"

In reply to UKC News:

AJPM is at http://www.ajpm-online.net/

Article is a load of nonce sense.

You need to have the number of participants in the "sport" at each time point that the number of injuries have been assessed and ideally the number of hours each participant has spent on rock to derive a rate per person hours for each time point in order to deduce whether there has been an increase.

Unfortunately it looks like a data dredging exercise where someone says "I know I've got access to all this information, I can publish a paper out of it." And they've trawled through and found something they can get written up and published to make their research group look competent (by virtue of having a greater number of published papers).

Its a shame journals accept such poor publications (having reviewed and rejected articles for The Lancet

In reply to UKC News:

Meant to link to http://www.ajpm-online.net/inpress where the article will presumably appear when its actually published (although I'd imagine you'll have to pay for it).

In reply to slacky:

Sounds like it'd be interesting for Mick to invite the authors to join in an on-line chat here, or perhaps a more formal interview.

At the least, it might be worth pointing the Research Group PR managers at this thread.

Y
(Former statistical safety analyst for a British military aircraft, who doesn't like to see rigorous safety analysis being brought into disrepute..)

In reply to Yanchik:

Beg pardon, I'm repeating myself. Must try more coffee with my whisky...

In reply to Yanchik:

I'll drop them an email.

In reply to Mick Ryan - UKClimbing.com:

Excellent ! It's all to win - some interesting statistically-based safety insights or the chance to gloat at a public humiliation...

Happy to take a look at the paper and put a couple of questions together for an interviewer if that's the direction it takes.

Y

In reply to Mick Ryan - UKClimbing.com:

By the way Mick, I wasn't criticising you for sticking this article on news (and I don't think anyone else was either).

But when you look in to what this woman has published recently from the same database and over widely varying subject areas, it looks like very many publications for very little research, thus adding up to very little credibility.

In reply to Yanchik:

My guess would be that she'll only attempt to defend herself in her own research community or the popular press. The opinions of a bunch of Brit climbers aren't going to affect her grant proposals or her publication scores.

That said, I hope she does and I don't think that we should shoot her down if she's good enough to come and properly defend her conclusions.

In reply to Hugh Cottam:

I agree with you. And, actually, I wasn't particularly critical of her work, but of the press release which fails to describe it - I would be delighted to see if she's wrung some useful conclusions out of a (probably) sloppy dataset.

However - research isn't done in a vacuum, divorced from reality. She appears to be attempting to describe our sport. The description should be able to meet our scrutiny.

Y

In reply to Hugh Cottam:

I know
(my skin is thick anyway and I rarely take offence)

I picked it up because I knew, with the talent who read this forum, there would be some great analysis.

In my opinion, this is study could have serious consequences for climbing, not just in the USA.

It's already been picked up by several nationals in the USA.

Perception is a powerful thing, as are headlines.

I've emailed the PR people at the Nationwide Children's Hospital in Ohio.

"It has been estimated that rock climbing is now enjoyed by more than 9 million people in the U.S. each year."

What a crock that figure is.

Try 200,000 max (in the USA) - 30-40,000 in the UK.

Mick

In reply to Mick Ryan - UKClimbing.com: You can email her direct as well.

In reply to Mick Ryan - UKClimbing.com:

>>>(my skin is thick anyway and I rarely take offence)

Well you say that. I'll have to see the stats.

>>>In my opinion, this is study could have serious consequences for climbing, not just in the USA.


Which is why it's so worrying, because it looks so poorly done. She's done exactly the same article on Computer Injuries using the same database. This type of research stinks because it's bad science. What's more she knows it, but doesn't care because she'll get the plaudits anyway.

In reply to Hugh Cottam:

As in any modern media I have a multi-faceted role, not just making sure that people know how popular we are, but even keeping my hand writing news reports: http://www.ukclimbing.com/news/item.php?id=48628

Which was fun to do.


I'm on it. Emailing Dougald at Climbing.com too.

Mick

In reply to UKC News: they didnt mention there is a 63% increase in people partaking in the activity of rock climbing in the US :P gotta love statistics -.-

In reply to UKC News:
I love reading shit journal papers. Can't wait to get home tonight and get stuck in.

In reply to Mick Ryan - UKClimbing.com:
I can not see why your estimate is any better or worse than the ones based in the report?

In 2003 it was considered there was 150,000 active climbers in the uk based on BMC membership and climbing wall stats. I also notice you have 48,000 current user id's on UKC. Ok I understand there are duplications etc but there are also many climbers not registered on UKC or with the BMC.

Without a good estimate of the volumes involved and the criteria you are using to classify what a “rock climber” is, any further analysis will quickly become wild speculation, but as you imply, perception can have more of an effect than reality. This is not a dig, but we all know stats can be read lots of ways depending on the results you want. I am sure however it would make be an interesting topic for UKC article to consider if you have access to the relevant data.

In reply to ClwydTraveller:

Well, Mick's livelihood depends on having a fair and widely-shared understanding of what a fair definition of what an "active climber" is, how many of them there are in different places, and ways to try to communicate with them.

I'm happy to rate that more highly than the credentials of the study.

Your figures are interesting. The implication would be that Mick's are an underestimate of 3-4 times, so total US participation might rise as high as 5-600 thousand. Nine million is an order of magnitude greater - to me, "a crock" seems adequately descriptive of that difference.

Nine million might be a fair guess at the "number of Americans who've been injured and possess gear that REI consider selling" - but if that's what the study shows, I certainly don't want anyone believing that the conclusions should apply to rock climbing.

Good to see the ideas and figures tested.

Y