Skip to main content
. 2019 Dec 13;2(12):e1917363. doi: 10.1001/jamanetworkopen.2019.17363

Table 3. Cerebrospinal Fluid Cutoffs for Maximal Discrimination Between Normal Cognition and Mild Cognitive Impairment by Race Using Receiver Operating Characteristic Curve Analyses.

Race by CSF Biomarker Cutoff (95% CI)a Sensitivity Specificity PPV NPV
African American
Aβ1-42 level, pg/mL 208 (126-321) 0.39 0.74 0.67 0.48
Tau level, pg/mL 51 (31-59) 0.39 0.82 0.74 0.51
Tau to Aβ1-42 ratio 0.17 (0.13-0.46) 0.5 0.66 0.66 0.5
pTau181 level, pg/mL 12 (12-19) 0.62 0.62 0.68 0.55
pTau181 to Aβ1-42 ratio 0.05 (0.03-0.12) 0.44 0.70 0.66 0.49
White
Aβ1-42 level, pg/mL 197 (183-245) 0.55 0.85 0.77 0.68
Tau level, pg/mL 59 (56-92) 0.66 0.72 0.68 0.7
Tau to Aβ1-42 ratio 0.23 (0.22-0.33) 0.74 0.72 0.70 0.75
pTau181 level, pg/mL 20 (12-27) 0.57 0.78 0.69 0.66
pTau181 to Aβ1-42 ratio 0.05 (0.05-0.13) 0.78 0.65 0.67 0.77

Abbreviations: Aβ1-42, β-amyloid 1-42; CSF, cerebrospinal fluid; NPV, negative predictive value; PPV, positive predictive value; pTau181, phosphorylated tau 181.

a

Calculated using logistic regression with receiver operating characteristic curve analysis to maximize Youden J index (the largest vertical distance from the uninformative diagonal line). The 95% CIs are calculated using bootstrap with sampling (×1000). Covariate-adjusted cutoffs are provided in eTable in the Supplement.