Table 2.
Association between baseline diabetes status and changes in cognitive Z-scores.
| Z-scores | Diabetes status | Model 1 | Model 2 | ||
|---|---|---|---|---|---|
| β (95% CI) | P-value | β (95% CI) | P-value | ||
| GCF | No-Diabetes (n=1023) | Ref. | Ref. | ||
| Prediabetes (n=2429) | -0.04 (-0.10, 0.03) | 0.277 | -0.01 (-0.04, 0.03) | 0.756 | |
| <5y Diabetes (n=667) | -0.12 (-0.20, -0.03) | 0.008* | -0.04 (-0.09, 0.01) | 0.109 | |
| ≥5y Diabetes (n=684) | -0.27 (-0.36, -0.18) | <0.001* | -0.11 (-0.16, -0.06) | <0.001* | |
| MMSE | No-Diabetes (n=1187) | Ref. | Ref. | ||
| Prediabetes (n=2786) | -0.01 (-0.07, 0.05) | 0.749 | 0.01 (-0.05, 0.06) | 0.865 | |
| <5y Diabetes (n=847) | -0.08 (-0.16, 0.01) | 0.054 | -0.05 (-0.13, 0.03) | 0.209 | |
| ≥5y Diabetes (n=865) | -0.11 (-0.19, -0.02) | 0.011* | -0.06 (-0.14, 0.02) | 0.134 | |
| CDT | No-Diabetes (n=1189) | Ref. | Ref. | ||
| Prediabetes (n=2788) | 0.01 (-0.06, 0.07) | 0.874 | 0.01 (-0.05, 0.07) | 0.780 | |
| <5y Diabetes (n=846) | -0.01 (-0.09, 0.08) | 0.843 | 0.01 (-0.08, 0.09) | 0.847 | |
| ≥5y Diabetes (n=866) | -0.09 (-0.18, -0.01) | 0.031 | -0.06 (-0.14, 0.03) | 0.171 | |
| DST-f | No-Diabetes (n=1072) | Ref. | Ref. | ||
| Prediabetes (n=2526) | -0.03 (-0.10, 0.05) | 0.474 | -0.01 (-0.08, 0.06) | 0.725 | |
| <5y Diabetes (n=702) | -0.08 (-0.17, 0.01) | 0.087 | -0.06 (-0.15, 0.03) | 0.198 | |
| ≥5y Diabetes (n=716) | -0.12 (-0.21, -0.03) | 0.012* | -0.07 (-0.16, 0.02) | 0.126 | |
| DST-b | No-Diabetes (n=1072) | Ref. | Ref. | ||
| Prediabetes (n=2525) | -0.04 (-0.11, 0.03) | 0.293 | -0.02 (-0.09, 0.04) | 0.528 | |
| <5y Diabetes (n=702) | -0.07 (-0.16, 0.02) | 0.116 | -0.04 (-0.13, 0.04) | 0.349 | |
| ≥5y Diabetes (n=716) | -0.11 (-0.20, -0.02) | 0.014* | -0.05 (-0.14, 0.04) | 0.251 | |
| VFT-a | No-Diabetes (n=1226) | Ref. | Ref. | ||
| Prediabetes (n=2866) | -0.07 (-0.13, -0.01) | 0.033 | -0.05 (-0.11, 0.01) | 0.101 | |
| <5y Diabetes (n=870) | -0.14 (-0.22, -0.05) | 0.001* | -0.10 (-0.17, -0.02) | 0.018 | |
| ≥5y Diabetes (n=889) | -0.25 (-0.33, -0.16) | <0.001* | -0.18 (-0.26, -0.10) | <0.001* | |
| VFT-p | No-Diabetes (n=1227) | Ref. | Ref. | ||
| Prediabetes (n=2865) | -0.05 (-0.12, 0.02) | 0.149 | -0.03 (-0.09, 0.03) | 0.348 | |
| <5y Diabetes (n=870) | -0.13 (-0.21, -0.04) | 0.005* | -0.08 (-0.16, 0.01) | 0.060 | |
| ≥5y Diabetes (n=889) | -0.23 (-0.32, -0.14) | <0.001* | -0.15 (-0.23, -0.07) | <0.001* | |
| TMT-A§ | No-Diabetes (n=1226) | Ref. | Ref. | ||
| Prediabetes (n=2862) | -0.02 (-0.08, 0.04) | 0.512 | -0.03 (-0.09, 0.03) | 0.323 | |
| <5y Diabetes (n=869) | 0.08 (0.01, 0.16) | 0.037 | 0.05 (-0.02, 0.13) | 0.185 | |
| ≥5y Diabetes (n=886) | 0.20 (0.11, 0.29) | <0.001* | 0.15 (0.06, 0.23) | 0.001* | |
| TMT-B§ | No-Diabetes (n=1221) | Ref. | Ref. | ||
| Prediabetes (n=2859) | 0.01 (-0.05, 0.07) | 0.690 | 0.01 (-0.06, 0.06) | 0.994 | |
| <5y Diabetes (n=866) | 0.11 (0.03, 0.20) | 0.006* | 0.08 (0.01, 0.16) | 0.039 | |
| ≥5y Diabetes (n=883) | 0.24 (0.15, 0.32) | <0.001* | 0.17 (0.09, 0.25) | <0.001* | |
<5y diabetes, less than 5 years diabetes duration; ≥5y diabetes, more than 5 years diabetes duration; GCF, Global Cognitive Function; MMSE, Mini-Mental State Examination; CDT, Clock Drawing Test; DST-f, Digit Span Test forward section; DST-b, Digit Span Test backward section; VFT-a, Verbal Fluency Test animal category; VFT-p, Verbal Fluency Test letter “p”; TMT-A, Trail Making Test part A; TMT-B, Trail Making Test part B.
§ Inverse neuropsychological assessment score.
Model 1: adjusted for sex, age (in years), intervention group, and center size (<250; 250-300, 300-400; ≥400).
Model 2: further adjusted for baseline education level (primary school; secondary school; college), civil status (single, divorced or separated; married; widower), physical activity (MET min/week), smoking habits (smoker; former smoker; never smoker), alcohol intake (g/day, adding the quadratic term), 17-point Mediterranean diet score, BMI (kg/m²), hypertension (yes/no), hypercholesterolemia (yes/no), and depressive symptomatology (yes/no).
Beta coefficients were estimated using linear regression models with robust standard errors to account for intracluster correlations.
*Significant association after Benjamini-Hochberg correction.