Open access. Just reading it now…
1 Like
Horse/athlete combination factor: high dressage score.
Phew. I’m safe.
That is very interesting.
9 Likes
So, they don’t seem to state this explicitly, but I think they are going by penalties (so a high score is a worse score, not a better one). This is why I think so:
“Horse-athlete combinations who recorded a score in the dressage phase that was higher than 50 were at increased odds of falling during the cross-country phase compared with combinations who recorded a dressage score of 50 or less… These risk factors, in part at least, indirectly reflect horse and athlete skill—it can be concluded from these along with the risk factor of dressage score that less experienced/less skilled horses and athletes were more likely to fall/be unseated.” Here they seem to be equating “over 50” with “less skilled.”
IMO 50 suggests a pretty disobedient test so that result doesn’t seem too surprising. I’d love to see the same analysis with different cut-offs, like 20-29, 30-39, and 40-49.
8 Likes
Yes, this is how I read it. It was a bad joke on my part. 
1 Like
It’s interesting that a recent start appears protective for riders but risky for horses.
“Horses with more starts in the recent past could end up being overworked and tired, and more likely to make a mistake, refuse at a fence or otherwise unseat their rider. Horses could also be experiencing sub-clinical injury which may affect their performance—it has previously been demonstrated that athletes do not always recognise when their horses are experiencing pain-related gait abnormalities. Associations between a higher number of recent race starts and increased risk of deleterious outcomes have previously been demonstrated, linking the accumulation of high-speed exercise to increased risk of injury. On the other hand, athletes with fewer recent starts were more likely to be unseated, indicating that for athletes it is more important to be well-practiced at competing.”
2 Likes
Sorry!! I interpreted the joke as “My dressage scores aren’t great so I’m safe!” 
1 Like
Perhaps the results should have been measuring the depth of pockets behind the riders (either as sponsors/owners/access to large-turnout style accomodation as opposed to expensive boarding eg)—those with multiple mounts at or near the level will not have to over-compete any one of them to gain the necessary mileage for themselves.
Well, those models may have some issues.
I don’t know the rules but isn’t length of course defined by level? If so the multi collinearity could a problem.
The variable on last start is perhaps not the best unless you control for seasonality AND the level on the recent start. I would expect it’s different for a start at a new level early on vs a start at a new level mid or late season. I don’t know if the way they bucketed the data on starts captures that. Is 180 days a number that would separate seasons. I don’t think so.
I also don’t get why gender of the horse would matter. What’s the hypothesis there? If it’s in the model because it’s a variable they had, that’s a red flag for a junky ad hoc model.
And they seem to have no data on weather or location on course of the fall. Late? Fatigue a factor. Early? Training more likely a factor.
I dunno how illuminating this is really.
1 Like
Long term primary data is available on the FEI and BE websites. Other nations don’t seem to publish safety data but all contribute to the FEI.
You will find many reports available on the FEI website;
Unfortunately the most important factor (that residing within the human head) is not part of the research 
1 Like
Im going to have to sit down and read through this…but today is not that day hahaha
Any other good tidbits in there?
The main thing to remember is that it was studying FEI starts.
In some parts of the world like Canada, it’s possible a combination is doing only a handful, or fewer, FEI starts per year. Therefore the whole picture can be skewed. Important to consider also the national HT, the schooling, and other factors like, did they travel 3 or 4 days to get to the FEI competition?
1 Like
The paper was specifically about FEI. I really do think national bodies should be publishing their own data at the end of each season. And no one is looking at schooling because that is outside organised competition.
3 Likes
I wasn’t suggesting that, precisely. I was saying that the results of this study may not hold a lot of value for a North American audience outside the very top competitors.
If you are only able to do one or two FEI events a year due to geography, you may or may not be able to compensate. Some people do lots of XC schooling, have great coaching, and do well at national HT on their way to great FEI results.
Others may not.
But either way analyzing their results is a far different beast than Europeans who often run CCI-shorts rather than national HT. They can re-route so easily to another FEI event. Maybe a fall 2 weeks after a unsuccessful FEI shows that “too close” competitions are riskier. That means nothing to those of us who have FEI events a few times a year, if you want to travel 1-4 days. Maybe it’s riskier the further apart events are if you’re in this continent… I’m not sure these stats are informative across all situations.
