Posted by rseiter on June 10, 2014
As part of trying to work out my recovery protocol I did some statistical analysis of the AchillesBlog recovery data and thought others might be interested (I can put the R code on GitHub if anyone wants to see it).
The current analysis is a bit rough at the moment (it contains my exploratory data analysis right now), but here are some results I found intriguing.
Note that it’s probably not good to focus on the 2014 results since they will be skewed by missing longer duration responses for recent injuries.
Looking at “time until” status by year it seems that time until PWB, FWB, and PT all show a decreasing trend by year, but time until Two Shoes is relatively constant. Here are PWB and TwoShoes, the others are at the end of the linked HTML (along with the numerical summary statistics).
It’s too bad the recovery data does not include op/non-op status. It would be interesting to see if the decrease is driven by non-op due to the recent UWO, Exeter, etc. studies (and to see if the op/non-op balance has changed over the years).