[BmoreKen]

As I lay here on my back crippled from a severe low back strain, I was searching the web for answers to my current condition and found this interesting article that looks to be recently published. The author investigated the incidence of overuse injuries among mountain bikers, and classified results according to a variety of terms including age, gender, riding frequency, suspension (rigid, front, full), gears (SS vs gears), and types of injuries.

http://www.mdpi.com/2075-4663/2/1/1/pdf

It looks like the most definitive conclusion, arguably, is that riding both SS and gears may predispose one to injuries given that the muscles used for either endeavor are not as fully developed as for someone who rides exclusively one or the other.

Thoughts?

Sent from my iPhone using Tapatalk

Sent from my iPhone using Tapatalk

 

[kjlued]

Hogwash

Too many factors involved to blame it on mixed riding however I would blame over developing one set of muscles without working out supporting muscles to be a higher factor in injuries than switching riding styles.

 

[BmoreKen]

Hogwash

Too many factors involved to blame it on mixed riding however I would blame over developing one set of muscles without working out supporting muscles to be a higher factor in injuries than switching riding styles.

I agree there are more variables, including age. Really tough to prove any link from such a small and always changing population.

Sent from my iPhone using Tapatalk

 

[kjlued]

I wouldn't call age a variable as much as fitness levels. I have seen people in their 60's and 70's in better shape than those in their 20's.

If you ride a gear bike, and just spin everywhere and then hop on a SS with a tall gear and try and climb with it, you will probably blow a knee out. If you swap back and forth and use the right gear for your fitness level on a SS you will be fine.

 

[BmoreKen]

I wouldn't call age a variable as much as fitness levels. I have seen people in their 60's and 70's in better shape than those in their 20's.

If you ride a gear bike, and just spin everywhere and then hop on a SS with a tall gear and try and climb with it, you will probably blow a knee out. If you swap back and forth and use the right gear for your fitness level on a SS you will be fine.

Yeah but how many 70 year olds are riding compared with 40 year olds?

Sent from my iPhone using Tapatalk

 

[kjlued]

You would be surprised.

Sure no where near as many people in their 70's as in their 40's riding around here but there are lots of people here still riding in their 60's and 70's. Some of them are beasts too. Most of them road but I see some older guys on the trails too.

 

[pwrtrainer]

I read this article. I wasn't too impressed when I read it at the beginning of the year. Lots of very bold statements and no reference. Lots of post hoc error. It gives me some ideas but the conclusions are too lofty. And the school is my alma mater so that is with my bias.

 

[BmoreKen]

heh, maybe all he/she did is capture who is more likely to report an injury...

 

[ca_rider]

The article focuses on overuse injuries but does not relate injury statistics to riding time with respect to their primary conclusion.

I would suggest the authors fit a curve relating (frequency of each injury) to (average periodic riding time) for each of the three groups (ss, ss/mg, mg). A comparison of the slope of the fits would be a much better metric to gauge the effects discussed.

It is trivial to call the conclusion of this paper in question with the following question: Did the ss/mg riders sustain more injuries because they ride more often? Perhaps this group rides more often, which is why they have two bikes.

The authors do include some metrics (Table 3) indicating that ride frequency increases risk but do not include this consideration in the results leading to their primary conclusion.

 

[pwrtrainer]

The article focuses on overuse injuries but does not relate injury statistics to riding time with respect to their primary conclusion.

I would suggest the authors fit a curve relating (frequency of each injury) to (average periodic riding time) for each of the three groups (ss, ss/mg, mg). A comparison of the slope of the fits would be a much better metric to gauge the effects discussed.

It is trivial to call the conclusion of this paper in question with the following question: Did the ss/mg riders sustain more injuries because they ride more often? Perhaps this group rides more often, which is why they have two bikes.

The authors do include some metrics (Table 3) indicating that ride frequency increases risk but do not include this consideration in the results leading to their primary conclusion.

exactly. Post hoc error. Love how prevalent it is in research.

 

[ca_rider]

I am curious about the authors' perspective and just sent them a formal communication. If they reply, I'll follow up here.

With due respect to the authors, they seem to be focused in the appropriate fields (two are Physical Therapists, one has a Masters Degree in Statistics). In their defense, they indicate a relationship between ss/mg riders with more succeptibility without digging deeper into the cause. The cause may simply be that these riders rode more, which is a relationship that exceeds the scope of the article. However I am not quite settled how this relates to theme of the paper.

 

[ca_rider]

Communication with the authors indicates that they included riding_time as an independent factor in their logistic regression model. As such, effects of riding_time should be separated from other effects in a least-squares sense.

However, page 5 indicates "a logistic regression model was constructed to determine associations between possible explanatory variables and reported overuse injuries. Included in this model were main effects of gender, riding volume, suspension type, and gear type." so this inclusion was not obvious to me. The paragraph indicates that riding_time is categorical, but does not indicate that it was included.

For those interpreting the data, it is useful to know they used a multivariate regression including more variables than seem to be indicated in the model descripton.