Table 2. Association Between Duration Played and CTE Neuropathology.

Factor	CTE neuropathologya	P value
Duration played b
CTE diagnosis, OR per 1-y duration (95% CI)	1.34 (1.15 to 1.55)	<.001
Cumulative ptau burden, β, SD per 1-y duration (95% CI)	0.037 (0.017 to 0.057)	<.001
Duration played and enforcer status
CTE diagnosis
OR per 1-y duration (95% CI)	1.30 (1.12 to 1.52)	<.001
Enforcer status, OR (95% CI)	2.03 (0.39 to 10.62)	.40
Cumulative ptau burden
β, SD per 1-y duration (95% CI)	0.029 (0.01 to 0.05)	.01
Enforcer status, β (95% CI), SD	0.35 (−0.05 to 0.74)	.08
Simulation assessing association between duration played and CTE neuropathology c
CTE diagnosis, median OR per 1-y duration (full range)	1.34 (1.29 to 1.40)	NA
Cumulative ptau burden, β, median SD per 1-y duration (full range)	0.039 (0.026 to 0.046)	NA

Abbreviations: CTE, chronic traumatic encephalopathy; NA, not applicable; OR, odds ratio; ptau, phosphorylated tau.

^a All models were adjusted for age at death, non–ice hockey contact sport play, age of first exposure to ice hockey play, position of play, coded vs not coded as forward (wingers and centers), and total concussions.

^b Separate models were run for each outcome. The sample size for all analyses is 77 donors (42 donors with and 35 donors without CTE).

^c For both simulations, median OR magnitudes were similar to those for the actual brain bank sample and the full range of association magnitudes were never null (ie, OR = 1 or SD increase = 0). Simulation assesses ORs with a range of nonnegative values for selection parameters for duration played (β_D; ie, log odds of brain donation for each additional year played when CTE pathology is absent), CTE diagnosis (β_C; ie, log odds of brain donation when CTE pathology is present compared with absent when duration played is approaching zero), and duration played × CTE diagnosis cross product (β_DC; ie, additional log odds of brain donation for each additional year played beyond β_D, when CTE pathology is present).