The prevalence and probability of elevated cortical amyloid according to surgical hospitalization. (A) The prevalence of elevated cortical amyloid levels among participants with previous surgical hospitalization and among each nonsurgery comparison group. Of note, the prevalence of amyloid-positive status among the combined group of participants without surgery was 46%. The P values were calculated using logistic regression models that adjusted for demographic factors, cardiovascular risk factors, and prevalent disease. (B) The estimated adjusted probability of elevated late-life (visit 5) brain amyloid according to the number of surgical hospitalizations. The values represent estimates from fully adjusted logistic regression models. Of the 280 participants included in this analysis, 54% (n = 151), 28% (n = 79), and 18% (n = 50) of participants had 0, 1, and at least 2 surgeries, respectively; 60% (n = 169), 25% (n = 69), and 15% (n = 42) of participants had 0, 1, and at least 2 moderate/high-risk surgeries, respectively. The test for differences among the all surgery groups and the moderate/high-risk surgery groups were nonsignificant with P values of P = 0.123 and P = 0.089, respectively. *P < .05 compared with the no-surgery group.