At baseline, mean age was 74.0 (3.0) years, 550 (55.2%) were female, and 538 (54.0%) were black. The mean (SD) β-amyloid 42 level of the 997 participants was 33.9 (9.6) pg/ml, of β-amyloid 40 was 191.6 (50.6) pg/ml, and of β-amyloid 42/40 was 0.19 (0.17). Participants in the low β-amyloid 42/40 tertile were more likely to be black (p=0.008), have diabetes (p=0.04), have lower literacy (p=0.03), smoke currently (p=0.05), and have an APOE e4 allele (p<0.001) (). Those in the low tertile tended to be less educated (p=0.06) but did not differ on other baseline characteristics.
In unadjusted mixed effects repeated measures models, β-amyloid 42/40 was not associated with baseline 3MS score (). However, low β-amyloid 42/40 was associated with greater cognitive decline over nine years (low tertile −6.59 (95% confidence interval −5.21 to −7.67) points, mid −6.16 −(4.92–7.32) and high −3.60 −(2.27–4.73), p-value<0.001) (). After adjustment for age, race, education, diabetes, smoking and APOE e4, the results remained statistically significant (−6.38 −(5.15–7.61) for low, −6.09 −(4.89–7.29) for mid and −3.44 −(2.21–4.67) for high tertile, p=0.02). There was also a significant association between plasma β-amyloid 42 and cognitive decline (low tertile −6.96 −(5.53–8.03), mid −5.14 −(3.94–6.34) and high −4.30 −(3.03–5.47), p=0.007), and multivariate adjustment for age, race, education, diabetes, smoking, and APOE e4 did not appreciably change these results (low tertile −6.70 −(5.45–7.95), mid −5.03 −(3.81–6.25) and high −4.29 −(3.06–5.52)). There was no association between plasma β-amyloid 40 and baseline cognitive function or decline. When β-amyloid 42/40 was analyzed as a continuous predictor, the β coefficient for 9-year cognitive decline in multivariate adjusted models is 0.68 per standard deviation of β-amyloid 42/40, p=0.03.
Mean (95% confidence interval) Modified Mini-Mental State Exam (3MS) score by plasma β-amyloid 42 and β-amyloid 42/40 tertile.
In multivariate adjusted models, cognitive reserve measures modified the association between β-amyloid 42/40 level and cognitive decline. Older adults with low reserve (as measured by education <HS or literacy ≤ 6th grade) had an even greater association with β-amyloid 42/40 level, whereas those with high reserve had less association (p for interaction <0.05 for both measures on 3MS decline) (). For example, among participants with education <HS, the 9-year decline on the 3MS was −8.94 −(6.94–10.94) for low β-amyloid 42/40 tertile compared with −4.45 −(2.31–6.59) for the high tertile, but among those with education ≥HS, the 9-year decline for low tertile was −4.60 −(3.07–6.13) and for high was −2.88 −(1.41–4.35) (p for interaction = 0.004).
Figure 1 a. Multivariate mixed effects model with adjusted mean 3MS score over years of follow-up by education level and β-amyloid 42/40 high and low tertile. Models are adjusted for age, race, diabetes, smoking and APOE e4; the interaction term is between (more ...)
APOE e4 status also modified the association between β-amyloid 42/40 level and cognitive decline, such that older adults with an e4 allele had an even greater association with β-amyloid 42/40 level. Among participants with an e4 allele, the 9-year decline on the 3MS was −7.75 −(5.75–9.75) for low tertile compared with −5.00 −(2.37–7.62) for high tertile but among those with no e4, decline was −5.50 −(3.95–7.05) for the low and −3.03 −(1.64–4.42) for the high tertile (p for interaction = 0.02).
Over the course of the follow-up, 72 participants had incident dementia and we repeated the analyses after excluding these elders. The association between plasma β-amyloid 42/40 tertile and 9-year cognitive decline remained statistically significant (low tertile −5.49 −(4.29–6.69), mid −5.26 −(4.10–6.42) and high −3.16 −(1.98–4.34), p=0.002). The interaction between education plasma β-amyloid 42/40 and cognitive decline remained statistically or borderline significant (p=0.03) whereas the interaction with literacy level was no longer significant (p=0.08) The interaction between APOE e4, β-amyloid 42/40, and cognitive decline also became non significant (p=0.43).