Table 3.
Top 10 adjusted associations from model C, small sMRI regions vs. visit 1 exposures: serum 25(OH)D, folate and cobalamin (overall and stratified analysis) with uncorrected P < 0.05: ordinary least square brain scan-wide analyses on HANDLS 2004–2009 and HANDLS-SCAN 2011–2015a.
| Outcome (vscan) | Outcome description | Exposure (v1) | Stratum | (N) | β | (SE) | P | Standardized Beta (b) | q-value | Passes FW Bonferroni correction | Standardized Beta (b): SAb | Puncorr: SA |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OVERALL | ||||||||||||
| Left_OCP_occipital_pole | Left occipital pole | 25(OH)D | __ | (186) | +15.70 | (3.83) | 6.3e-05 | +0.31 | 0.026 | Yes | +0.27 | < 0.001 |
| occipital_lobe_WM_left | Occipital lobe, white matter, left | 25(OH)D | __ | (186) | +76.8 | (20.7) | 2.9e-04 | +0.26 | 0.061d | No | +0.26 | < 0.001 |
| Right_PoG_post-central_gyrus | Post-central gyrus, right | 25(OH)D | __ | (186) | +34.8 | (9.7) | 4.3e-04 | +0.27 | 0.061d | No | +0.27 | 0.001 |
| parietal_lobe_WM_right | Parietal lobe, white matter, right | 25(OH)D | __ | (186) | +127.9 | (38.1) | 9.8e-04 | +0.24 | 0.10d | No | +0.23 | 0.002 |
| Left_PoG_post-central_gyrus | Post-central gyrus, left | 25(OH)D | __ | (186) | +34.1 | (10.4) | 1.3e-03 | +0.25 | 0.11d | No | +0.25 | 0.002 |
| Right_TrIFG_triangular_part_of_t | Triangular part of the inferior frontal gyrus, right | B-12 | __ | (240) | +0.45 | (0.14) | 2.2e-03 | +0.20 | 0.13 | No | +0.19 | 0.017 |
| parietal_lobe_WM_left | Parietal lobe, white matter, left | 25(OH)D | __ | (186) | +123.4 | (40.4) | 3.1e-03 | +0.22 | 0.13d | No | +0.21 | 0.007 |
| occipital_lobe_WM_right | Occipital lobe, white matter, right | 25(OH)D | __ | (186) | +63.6 | (21.1) | 3.0e-03 | +0.21 | 0.13d | No | +0.21 | 0.005 |
| Right_TMP_temporal_pole | Right temporal pole | FOL | __ | (240) | −35.5 | (11.9) | 2.7e-03 | −0.19 | 0.13 | No | −0.22 | 0.010 |
| Anterior insula, right | Right_AIns_anterior_insula | B-12 | __ | (240) | +0.36 | (0.12) | 3.2e-03 | +0.17 | 0.13 | No | +0.13 | 0.071 |
| STRATIFIED | ||||||||||||
| Left_OCP_occipital_pole | Left occipital pole | 25(OH)D | AP | (132) | +19.0c | (4.3) | 2.0e-05 | +0.35 | 0.07 | Yes | +0.32 | < 0.001 |
| Right_TMP_temporal_pole | Right temporal pole | FOL | Whites | (141) | −63.9c | (15.2) | 4.8e-05 | −0.34 | 0.08d | No | −0.42 | < 0.001 |
| Left_OCP_occipital_pole | Left occipital pole | 25(OH)D | Men | (87) | +24.0 | (5.8) | 8.0e-05 | +0.45 | 0.09d | No | +0.46 | < 0.001 |
| Left_OCP_occipital_pole | Left occipital pole | 25(OH)D | Whites | (109) | +17.7 | (4.5) | 1.6e-04 | +0.33 | 0.11d | No | +0.31 | 0.001 |
| Right_PoG_post-central_gyrus | Right post-central gyrus | 25(OH)D | Men | (87) | +64.4c | (16.6) | 1.6e-04 | +0.43 | 0.13d | No | +0.47 | 0.001 |
| Parietal_lobe_WM_right | Right parietal lobe, White matter | 25(OH)D | Men | (87) | +242.6c | (63.4) | 2.6e-04 | +0.45 | 0.13d | No | +0.46 | 0.001 |
| occipital_lobe_WM_left | Occipital lobe, white matter, left | 25(OH)D | >50 | (80) | +107.7 | (28.0) | 3.4e-04 | +0.37 | 0.14d | No | +0.28 | 0.004 |
| parietal_lobe_WM_left | Parietal lobe, white matter, left | 25(OH)D | >50 | (80) | +201.8c | (53.6) | 3.4e-04 | +0.39 | 0.14d | No | +0.34 | 0.003 |
| occipital_lobe_WM_right | Occipital lobe, white matter, right | 25(OH)D | Men | (87) | +132.3c | (35.9) | 4.1e-04 | +0.43 | 0.14d | No | +0.44 | 0.001 |
| Right_PHG_parahippocampal_gyrus | Right parahippocampal gyrus | FOL | Whites | (141) | −20.6c | (5.7) | 4.2e-04 | −0.27 | 0.14 | No | −0.38 | < 0.001 |
25(OH)D, 25-hydroxyvitamin D; AP, Above poverty; B-12, serum cobalamin (vitamin B-12); FDR, False Discovery Rate; FOL, serum folate; FWER, FamilyWise Error Rate; GM, Gray Matter; SA, Sensitivity Analysis; SE, Standard Error; WM, White Matter.
Values are adjusted linear regression coefficients β with associated SE, standardized beta, uncorrected p-values, corrected q-values (false discovery rate) and results of sensitivity analysis. (N) is the sample size in each analysis. Bolded rows correspond to statistically significant associations after correction for multiple testing, FWER < 0.05.
Based on a sensitivity analysis further adjusting for selected socio-demographic, lifestyle and health-related factors after screening using machine learning techniques (see Supplemental Methods 2). Note that for visit 1 25(OH)D, no additional covariates were selected. For Folate and B-12 a reduced set of additional covariates were included and are listed in Supplemental Methods 2.
P < 0.10 for null hypothesis that exposure × stratifying variable 2-way interaction term is =0 in the unstratified model with exposure and socio-demographic factors included as main effects.
Finding considered a trend for passing FDR q-value correction at type I error of 0.10 per vitamin, model and stratification status while failing the FWER criterion, due to a standardized effect size (in absolute value) ≥0.20.