Table A1.
Association Between Folate, Vitamin B12, and Homocysteine and Prevalent Cognitive Impairment at Baseline and Incident Cognitive Impairment After 5 Years (SMMSE≤25) in the Newcastle 85+ Study
| Prevalent Cognitive Impairment |
Incident Cognitive Impairment |
|||||||
|---|---|---|---|---|---|---|---|---|
| Model 1 |
Model 2 |
Model 1 |
Model 2 |
|||||
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| RBC folate, nmol/L | ||||||||
| Q2 (612–870) | 0.77 (0.44–1.33) | .35 | 0.73 (0.40–1.31) | .29 | 1.68 (0.68–4.14) | .26 | 1.68 (0.61–4.65) | .31 |
| Q3 (870–1280) | 0.74 (0.43–1.28) | .29 | 0.71 (0.38–1.30) | .26 | 2.40 (1.02–5.65) | .05 | 2.59 (0.96–7.02) | .06 |
| Q4 (>1280) | 0.84 (0.49–1.45) | .53 | 0.73 (0.40–1.34) | .31 | 1.10 (0.42–2.87) | .84 | 0.84 (0.27–2.67) | .88 |
| Plasma vitamin B12, pmol/L | ||||||||
| Q2 (170–232) | 0.80 (0.46–1.40) | .44 | 0.64 (0.35–1.18) | .16 | 1.15 (0.52–2.53) | .73 | 1.41 (0.56–3.53) | .46 |
| Q3 (232–325) | 1.06 (0.62–1.81) | .84 | 1.03 (0.57–1.87) | .93 | 0.64 (0.26–1.54) | .32 | 0.61 (0.23–1.66) | .34 |
| Q4 (>325) | 0.89 (0.51–1.54) | .66 | 0.67 (0.36–1.23) | .19 | 0.85 (0.36–2.03) | .72 | 1.12 (0.40–4.11) | .84 |
| Total homocysteine, μmol/L | ||||||||
| Q2 (13.5–16.7) | 0.99 (0.56–1.75) | .96 | 1.05 (0.57–1.96) | .87 | 1.12 (0.50–2.54) | .79 | 1.71 (0.68–4.30) | .26 |
| Q3 (16.7–21.4) | 1.18 (0.67–2.09) | .57 | 1.40 (0.75–2.60) | .29 | 0.79 (0.34–1.88) | .60 | 0.97 (0.37–2.56) | .96 |
| Q4 (>21.4) | 1.55 (0.89–2.69) | .13 | 2.15 (1.12–4.11) | .02 | 1.10 (0.46–2.64) | .83 | 1.81 (0.63–5.25) | .27 |
Quartile 1 (<612 nmol/L RBC folate, <170 pmol/L plasma vitamin B12, and <13.5 tHcy, respectively) was used as the reference category (1.00) for all models. Model 1 is unadjusted, Model 2 is adjusted for alcohol intake, smoking status, APOE genotype (rs429358 and rs7412), sex, education, BMI, depression, hypertension, diabetes type 1 and 2, history of cardiovascular diseases, and physical activity. RBC folate and plasma vitamin B12 models were additionally adjusted for tHcy and the homocysteine model for renal impairment.