HJAR Sep/Oct 2023

HEALTHCARE JOURNAL OF ARKANSAS I  SEP / OCT 2023 31 relationship observed between NFT burden, but not CTE status or severity, for running backs compared to offensive linemen. These results provide additional evi- dence that repeated nonconcussive inju- ries are associated with CTE pathology. This is in contrast to the emphasis on concus- sions that is often discussed in the medical and lay literature 48 . Further, these results suggest that models incorporating inten- sity of impacts (i.e., linear and rotational acceleration) have better model fit and are better at predicting CTE status and sever- ity than models incorporating duration of play or number of hits to the head alone. These results, if validated, could be used to identify changes to policy or gameplay that might limit CTE risk by decreasing cumula- tive exposure, such as by limiting duration of exposure, the number of exposures, and the magnitude of those exposures. This study also validates the use of a PEM to characterize the aspects of RHI exposure experienced by football players. Other out- comes besides CTE believed to be associ- ated with RHI exposure, such as measures of diffuse axonal injury 49 , neuroinflammatory markers 50,51 , or other neuropathologic pro- cesses 52 , could be evaluated in the context of exposures derived from the PEM to better understand the characteristics of the expo- sure most responsible for these changes. Several biomechanical factors may explain why models testing the association between cumulative acceleration and CTE pathology have better fit thanmodels incor- porating the number of head impacts alone. Increasing linear forces are associated with focal injuries 53 ; greater forces may therefore result in more areas of localized damage 54 . Furthermore, rotational forces may result in increased shear-induced damage, particu- larly around small blood vessels and at the depths of the sulci 49 . This study utilized a PEM generated solely from available helmet accelerome- ter literature. Helmet accelerometer stud- ies typically report results pertaining to head impacts greater than 10–15 g 35 . These thresholds were chosen to exclude inciden- tal movements and jostling between the hel- met and head 55 , but there is no evidence that these thresholds are most relevant to CTE pathology 56 . Future work should explore different thresholds for counting hits, to fur- ther characterize the nature of head impacts most responsible for CTE pathology. The position an athlete plays serves as a proxy for the RHI exposure experi- enced. The fact that position at the highest level was not associated with CTE status, CTE severity, or NFT burden (the latter for all positions but running backs), whereas cumulative measures incorporating posi- tion were, suggests that one single posi- tion inadequately represents an athlete’s