Development of an Injury Burden Prediction Model in Professional Baseball Pitchers.
Garrett Bullock, Charles Thigpen, Gary Collins and 4 others
PMID 36518836WHAT IT FOUND
A statistical model predicted days lost to arm injury in professional baseball pitchers with an error of 12 days.
It accurately identified high-risk pitchers but cannot reliably distinguish between minor injuries and no injury.
Key findings
01The prediction model had a root mean square error of 11.9 days, meaning actual days lost would typically fall within 12 days of the prediction.
02Predictions for low injury burdens (1 to 14 days) are unreliable because they fall within the model's error margin.
03Including both modifiable and non-modifiable factors improved prediction performance compared to using only modifiable factors.
STILL TO COME
How it was doneWhat they foundWhat it means for PTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The model was developed using data from pitchers in a single Major League Baseball organization, so its results may not apply to pitchers in other organizations or at different levels of play. The model was not externally validated, meaning it has not been tested on a completely separate group of pitchers to confirm its accuracy. Predictions for low injury burdens (1 to 14 days) are unreliable because they fall within the model's error margin. Physical measurements were taken only before the season, so changes in range of motion or strength during the season were not accounted for in the prediction. Injury history was limited to professional seasons and prior orthopaedic surgeries, potentially missing relevant pre-professional injuries.
Declared interests
The authors declared no conflicts of interest. One author was supported by the NIHR Biomedical Research Centre, Oxford, and Cancer Research UK.
The easy way to misread this
Do not interpret a prediction of 1 to 14 days lost as evidence that a pitcher will sustain a minor injury. The model's error margin is 12 days, so a prediction in this range is statistically indistinguishable from zero days lost.