. Author manuscript; available in PMC: 2022 Jul 1.

Published in final edited form as: Eur J Nucl Med Mol Imaging. 2021 Jan 4;48(7):2259–2271. doi: 10.1007/s00259-020-05152-8

Table 2.

Comparison of subjects with concordant T-status across all modalities and subjects with discordant T-status in at least one modality pair. Mean ± SD is shown for continuous variables

	Concordant T− (n = 99)	Concordant T+ (n = 107)	Discordant T status (n = 145)	p value^*
Age	68.1 ± 8.3	68.4 ± 9.3	73.7 ± 8.5	p < .001^†
Sex (M, %)	39 (39%)	39 (36%)	78 (54%)	p = .01
Education	15.6 ± 3.9	15.4 ± 3.3	15.5 ± 3.6	p = .83
MMSE	28.3 ± 2.5	23.2 ± 5.0	26.9 ± 4.6	p <.001^†
Amyloid status (Aβ+, %)	14 (14%)	106 (99%)	76 (52%)	p <.001
CSF PTau181 (pg/mL)	16.1 ± 4.2	42.4 ± 16.6	23.2 ± 8.5	p <.001
FTP SUVR	1.14 ± 0.1	1.97 ± 0.5	1.28 ± 0.3	p < .001
Diagnosis (% Aβ+)	52 CN (15%)	9 CN (89%)	66 CN (48%)	p < .001
	23 MCI (17%)	38 MCI (100%)	45 MCI (42%)
	2 AD_c (0%)	59 AD_c (100%)	23 AD_c (96%)
	22 non–AD (9%)	1 non-AD (100%)	11 non-AD (27%)

p values shown for Kruskal-Wallis H test for continuous variables, and x² for nominal variables

^†

Bonferroni-corrected p values for pairwise comparisons are all < .001 except; T− vs discordant T for age (p = 1.0), and T– vs discordant T for MMSE (p = .06)