Table 2.
Intensive treatment | Standard treatment | – | – | – | |
---|---|---|---|---|---|
Adjudicated cognitive impairment (N = 8563, median follow-up = 5.1 years) | |||||
– | Cases per | Cases per | Hazard ratio | – | – |
Outcome | 1,000 person-years | 1,000 person-years | (95% CI)a | p value | Reference |
Probable dementia | 7.2 | 8.6 | 0.83 (0.67–1.04) | 0.10 | Williamson et al.15 |
Mild cognitive impairment | 14.6 | 18.3 | 0.81 (0.69–0.95) | 0.007 | |
Composite of MCI or probable dementia | 20.2 | 24.1 | 0.85 (0.74–0.97) | 0.01 | |
Cognitive decline outcomes (N = 2921, median follow-up = 4.1 years) | |||||
Yearly | Yearly | Difference | – | – | |
Outcome | Slope (95% CI)b | Slope (95% CI)b | (95% CI) | p value | Reference |
Memory domainc | −0.005 | −0.001 | −0.004 | 0.33 | Rapp et al.25 |
(−0.01 to 0.001) | (−0.006 to 0.005) | (−0.012 to 0.004) | – | ||
Processing domaind | −0.025 | −0.015 | −0.01 | 0.02 | |
(−0.03 to −0.019) | (−0.021 to −0.009) | (−0.017 to −0.002) | – | ||
MRI outcomes (baseline N = 670, median follow-up = 4.0 years) | |||||
Mean change | Mean change | Difference | – | – | |
Outcome | (95% CI)e | (95% CI)e | (95% CI) | p value | Reference |
White matter lesion volume, cm3 | 0.92 | 1.45 | −0.54 | <0.001 | Williamson et al. 201915; Nasrallah et al. 202128; Dolui et al. 2022.27 |
(0.69–1.14) | (1.21–1.70) | (-0.87 to −0.20) | – | ||
Total brain volume, cm3 | −30.60 | −26.9 | −3.7 | 0.006 | |
(−32.3 to −28.8) | (−28.8 to −24.9) | (−6.3 to −1.1) | – | ||
Hippocampal volume, cm3 | −0.06 | −0.02 | −0.033 | 0.03 | |
(−0.08 to −0.04) | (−0.05 to 0.00) | (−0.062 to −0.003) | – | ||
Whole brain CBF, mL/100 g/min | 1.46 | −0.84 | 2.3 | 0.02 | |
(0.08–2.83) | (−2.30 to 0.61) | (0.30–4.30) | – |
Intensive treatment versus standard treatment based on stratified Cox proportional hazards regression model.
Yearly slope assuming a linear trend based on a linear mixed model.
Includes the Logical Memory I and II, Modified Rey-Osterrieth Complex Figure (immediate recall), and the Hopkins Verbal Learning Test, Revised (delayed recall).
Includes the Trail Making Test (parts A and B) and Digit Symbol Coding.
For MRI outcomes, change estimates at 3.98 years post randomization based on a linear mixed model minimally adjusting for intracranial volume and days since randomization, including random effects for participant and imaging facility. Estimates for white matter lesion volume are based on a robust mixed model formulation given the skewed distribution of that outcome. Estimates for hippocampal volume and CBF also adjusted for age and sex.