Table 3.
Extended Cox Models, estimating hazards of progression to AD dementia as functions of the covariates and with neurochemically possible AD category as a time-dependent variable, split into “before 3 years follow-up” and “after 3 years follow-up”
| Covariate | M0 (n = 144) | M1 (n = 144) | M2 (n = 139) | M3 (n = 131) |
|---|---|---|---|---|
| Erlangen Score (reference category: Improbable AD) | ||||
| Possible AD | ||||
| Before 3-year follow-up | 7.67 (2.08–28.27)* | 7.75 (2.05–29.27)* | 7.43 (1.95–28.37)* | 5.68 (1.47–21.97)* |
| After 3-year follow-up | 1.04 (0.20–5.47) | 1.04 (0.20–5.50) | 0.94 (0.18–4.99) | 0.92 (0.17–4.91) |
| Probable AD | 12.0 (3.67–39.46)* | 12.3 (3.60–42.19)* | 10.8 (3.09–37.38)* | 8.09 (2.29–28.56)* |
| Age (years) | – | 1.00 (0.97–1.03) | 1.00 (0.97–1.03) | 1.00 (0.96–1.03) |
| Female gender | – | 1.10 (0.68–1.81) | 1.07 (0.66–1.75) | 1.05 (0.63–1.75) |
| APOE ε4 presence | – | – | 1.42 (0.86–2.36) | 1.48 (0.88–2.48) |
| MMSE | – | – | – | 0.95 (0.87–1.02) |
| Log likelihood | − 284.0 | − 283.9 | − 280.6 | − 264.5 |
Compare with the corresponding Kaplan–Meier curve (Fig. 1). Hazard ratios compared to reference category (neurochemically improbable, i.e. Erlangen Score = 0 or 1) presented together with the corresponding 95% confidence intervals
AD Alzheimer’s disease, APOE apolipoprotein E, M0 model with ES as the only covariate, M1 M0 supplemented with age and gender, M2 M1 supplemented with APOE genotype, M3 M2 supplemented with MMSE score, MMSE Mini Mental State Examination
*Statistical significance at p < 0.05 level, compared to the reference category