Table 4.
Cognitive test cutoffs at 1.5 SD from mean for cohorts A, B and E
| Cohort A cutoffs (95% CI) | Cohort B cutoffs (95% CI) | Cohort E cutoffs (95% CI) | |
|---|---|---|---|
| ADAS-delayed recall | 4.88 (4.59–5.16) | 4.38 (4.13–4.61) | 3.93 (3.63–4.20) |
| ADAS naming | 1.57 (1.42–1.71) | 1.44 (1.29–1.58) | 1.25 (1.01–1.48) |
| Animal Fluency | 13.5 (13.1–13.9) | 14.0 (13.6–14.4) | 15.0 (14.4–15.7) |
| AQT | 85.3 (83.6–86.9) | 83.5 (81.8–85.2) | 80.0 (77.8–82.1) |
| Stroop | 40.0 (39.0–40.9) | 39.0 (38.0–39.8) | 36.9 (35.1–38.6) |
| TMT A | 71.5 (69.2–73.9) | 71.3 (68.9–73.8) | 59.0 (56.5–61.2) |
| TMT B | 180.6 (172.1–189.0) | 176.0 (167.8–184.4) | 144.6 (135.5–152.5) |
| SDMT | 24.3 (23.7–25.0) | 25.1 (24.4–25.8) | 27.2 (26.1–28.3) |
Cutoffs (1.5 SD from mean) created from 500 bootstrap samples. We found significantly improved cutoffs in cohort E compared to cohort A for all tests apart from ADAS naming. We found significantly improved cutoffs in cohort E compared to the traditional method of creating robust norms (cohort B) for Animal Fluency, TMT A, TMT B, and SDMT, i.e., non-overlapping 95% CIs (presented in bold)