Table 5.
Item-Level Analysis of the rFSRP
| TMTA | TMTB | Digit Symbol | CDR Sum of Boxes | Log-WMH (quartiles) |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| rFSRP Items | β | b (95% CI) |
P | β | b (95% CI) |
P | β | b (95% CI) |
P | β | b (95% CI) |
P | OR | 95% CI | P |
| Age | 0.23 | 0.68 (0.64, 0.72) |
<0.01 | 0.24 | 2.41 (2.27, 2.54) |
<0.01 | −0.32 | −0.54 (−0.56, −0.51) |
<0.01 | 0.13 | 0.04 (0.03, 0.04) |
<0.01 | 1.11 | 1.04, 1.17 | <0.01 |
| Sex | −0.07 | −3.80 (−4.50, −3.09) |
<0.01 | −0.07 | −12.11 (−14.33, −9.89) |
<0.01 | 0.17 | 4.94 (4.54, 5.34) |
<0.01 | −0.12 | −0.57 (−0.63, −0.50) |
<0.01 | 1.16 | 0.89, 1.52 | 0.28 |
| Current smoking | 0.02 | 2.68 (0.95, 4.42) |
<0.01 | 0.02 | 7.31 (1.83, 12.78) |
0.01 | −0.03 | −2.14 (−3.13, −1.14) |
<0.01 | 0.02 | 0.23 (0.07, 0.39) |
0.01 | 2.27 | 1.21, 4.26 | 0.01 |
| CVD | 0.01 | 0.73 (−0.20, 1.66) |
0.13 | 0.004 | 0.86 (−2.08, 3.79) |
0.57 | −0.003 | −0.10 (−0.63, 0.42) |
0.70 | 0.02 | 0.11 (0.03, 0.20) |
0.01 | 1.29 | 0.97, 1.72 | 0.08 |
| AF | 0.004 | 0.44 (−0.94, 1.81) |
0.53 | 0.02 | 6.05 (1.73, 10.37) |
0.01 | −0.01 | −0.77 (−1.54, 0.01) |
0.05 | 0.001 | 0.01 (−0.12, 0.14) |
0.88 | 1.05 | 0.68, 1.62 | 0.82 |
| Anti-hypertensive | −0.001 | −0.06 (−0.78, 0.67) |
0.88 | −0.01 | −1.47 (−3.77, 0.83) |
0.21 | −0.02 | −0.44 (−0.86, −0.03) |
0.04 | −0.02 | −0.09 (−0.16, −0.02) |
0.01 | 1.31 | 1.03, 1.65 | 0.03 |
| SBP | 0.001 | 0.001 (−0.02, 0.02) |
0.94 | 0.03 | 0.13 (0.07, 0.19) |
<0.01 | −0.03 | −0.03 (−0.037, −0.015) |
<0.01 | 0.01 | 0.002 (0.000, 0.003) |
0.07 | 1.01 | 1.00, 1.02 | 0.001 |
| DM | 0.02 | 1.78 (0.72, 2.84) |
<0.01 | 0.03 | 7.31 (3.96, 10.66) |
<0.01 | −0.05 | −2.24 (−2.85, −1.63) |
<0.01 | 0.03 | 0.19 (0.09, 0.29) |
<0.01 | 1.48 | 1.09, 1.99 | 0.01 |
Multivariable linear regression examined the effect of the rFSRP items on TMTA, TMTB, Digit Symbol, and the CDR Sum of Boxes. Ordinal logistic regression tested the effect of the rFSRP items on quartile rankings of Log-WMH. Covariates included in the model were years of education, racial identity, and APOE e4. For models that examined log-transformed WMH (categorized into quartiles), age at MRI (as opposed to age at baseline) and estimated intracranial volume were also included. Higher scores are worse for TMTA, TMTB, CDR Sum of Boxes, and WMH; lower scores are worse for WAIS-R Digit Symbol. For sex, 1 = female and 0 = male; and 1 = present and 0 = absent for current cigarette smoking status, prevalent cardiovascular disease, atrial fibrillation, antihypertensive medication use, and diabetes.
Sample size for the neuropsychological tests and CDR Sum of Boxes was 19,309 with the exception of the WAIS-R Digit Symbol for which the sample was reduced to 16,447 due to missing data. Sample size for MRI indices was 1,196.
Abbreviations: β = standardized beta; b = unstandardized beta; OR = odds ratio; TMTA = Trail Making Test Part A; TMTB = Trail Making Test Part B; CDR = CDR Dementia Staging Instrument; CVD = prevalent cardiovascular disease defined by recent or active heart attack, congestive heart failure, and other cardiovascular events; AF = prevalent or past atrial fibrillation; SBP = systolic blood pressure; antihypertensive = current antihypertensive medication use defined by antihypertensive or blood pressure medication and antihypertensive combination therapy; DM = prevalent or past diabetes mellitus