Mediterranean, DASH, and MIND dietary patterns and cognitive function: the 2-year longitudinal changes in an older Spanish cohort
Abstract
Background and Aims: Plant-forward dietary patterns have been associated with cardiometabolic health benefits, which, in turn, have been related to cognitive performance with inconsistent findings. The objective of this study was to examine the relationship between baseline adherence to three a priori dietary patterns (Mediterranean, DASH, and MIND diets) with 2-year changes in cognitive performance in older adults with overweight or obesity and high cardiovascular disease risk. Methods: A prospective cohort analysis was conducted within the PREDIMED-Plus trial, involving 6,647 men and women aged 55–75 years with overweight or obesity and metabolic syndrome. Using a validated, semiquantitative 143-item food frequency questionnaire completed at baseline, the dietary pattern adherence scores were calculated. An extensive neuropsychological test battery was administered at baseline and 2-year follow-up. Multivariable-adjusted linear regression models were used to assess associations between 2-year changes in cognitive function z-scores across tertiles of baseline adherence to the a priori dietary patterns. Results: Adherence to the Mediterranean diet at baseline was associated with 2-year changes in the general cognitive screening Mini-Mental State Examination (MMSE, β: 0.070; 95% CI: 0.014, 0.175, P-trend = 0.011), and two executive function-related assessments: the Trail Making Tests Part A (TMT-A, β: −0.054; 95% CI: −0.110, − 0.002, P-trend = 0.047) and Part B (TMT-B, β: −0.079; 95% CI: −0.134, −0.024, P-trend = 0.004). Adherence to the MIND diet was associated with the backward recall Digit Span Test assessment of working memory (DST-B, β: 0.058; 95% CI: 0.002, 0.114, P-trend = 0.045). However, higher adherence to the DASH dietary pattern was not associated with better cognitive function over a period of 2 years. Conclusion: In older Spanish individuals with overweight or obesity and at high cardiovascular disease risk, higher baseline adherence to the Mediterranean dietary pattern may be associated with better cognitive performance than lower adherence over a period of 2 years.
Full text
Page 1 of 2 SUPPLEMENTAL TABLE Supplemental Table 1. Longitudinal 2-year analysis of multivariable adjusted mean change (β [95% CI]) in cognitive function assessments according to tertiles in level of adherence to each dietary pattern at baseline. Dietary Pattern MedDiet DASH MIND Neurological Assessment1 Lowest Adherence Modest Adherence Highest Adherence p for trend Lowest Adherence Modest Adherence Highest Adherence p for trend Lowest Adherence Modest Adherence Highest Adherence p for trend Score, median (range) 6 (1 to 7) 8 (8 to 9) 10 (10 to 14) 19 (8 to 21) 24 (22 to 26) 30 (27 to 38) 8 (2.5 to 8.5) 9 (9.0 to 9.5) 10.5 (10.0 to 13.5) N 1904 1790 914 1541 1578 1489 2058 1416 1134 GCF2 N 1663 1786 1159 1589 1573 1446 2058 1416 1134 Crude model 0 (ref.) 0.010 (-0.057, 0.077) 0.031 (-0.045, 0.107) 0.428 0 (ref.) -0.159 (-0.228, -0.091)a -0.317 (-0.388, -0.246)a <0.001 0 (ref.) -0.064 (-0.132, 0.005) -0.027 (-0.099, 0.045) 0.412 Model 1 0 (ref.) 0.034 (-0.002, 0.070) 0.038 (-0.004, 0.080) 0.059 0 (ref.) -0.044 (-0.081, -0.006)a -0.022 (-0.062, 0.018) 0.257 0 (ref.) -0.027 (-0.064, 0.010) 0.022 (-0.018, 0.062) 0.333 Model 2 0 (ref.) 0.031 (-0.005, 0.067) 0.033 (-0.009, 0.075) 0.095 0 (ref.) -0.042 (-0.080, -0.004)a -0.027 (-0.067, 0.014) 0.154 0 (ref.) -0.020 (-0.057, 0.016) 0.023 (-0.017, 0.063) 0.367 MMSE N 1969 2122 1365 1825 1866 1765 2421 1678 1357 Crude model 0 (ref.) 0.046 (-0.014, 0.106) 0.074 (0.008, 0.139)a 0.026 0 (ref.) -0.081 (-0.140, -0.021)a -0.178 (-0.241, -0.115)a <0.001 0 (ref.) 0.001 (-0.058, 0.061) 0.007 (-0.056, 0.070) 0.822 Model 1 0 (ref.) 0.059 (0.007, 0.111)a 0.077 (0.020, 0.133)a 0.006 0 (ref.) 0.007 (-0.045, 0.059) -0.003 (-0.059, 0.053) 0.906 0 (ref.) 0.034 (-0.018, 0.086) 0.036 (-0.017, 0.089) 0.173 Model 2 0 (ref.) 0.053 (0.002, 0.104)a 0.068 (0.017, 0.125)a 0.014 0 (ref.) 0.013 (-0.039, 0.065) -0.002 (-0.058, 0.054) 0.941 0 (ref.) 0.044 (-0.007, 0.095) 0.039 (-0.014, 0.092) 0.139 CDT N 1965 2128 1364 1822 1869.000 1766 2420 1675 1362.000 Crude model 0 (ref.) -0.030 (-0.091, 0.030) -0.001 (-0.068, 0.066) 0.879 0 (ref.) -0.098 (-0.160, -0.037)a -0.170 (-0.234, -0.107)a <0.001 0 (ref.) -0.028 (-0.090, 0.034) 0.0003 (-0.064, 0.065) 0.971 Model 1 0 (ref.) -0.009 (-0.064, 0.047) 0.021 (-0.042, 0.084) 0.557 0 (ref.) -0.031 (-0.089, 0.028) -0.030 (-0.090, 0.031) 0.336 0 (ref.) -0.008 (-0.066, 0.050) 0.028 (-0.032, 0.087) 0.376 Model 2 0 (ref.) -0.015 (-0.071, 0.040) 0.011 (-0.052, 0.074) 0.808 0 (ref.) -0.026 (-0.085, 0.032) -0.029 (-0.090, 0.032) 0.351 0 (ref.) 0.002 (-0.056, 0.060) 0.030 (-0.030, 0.090) 0.329 VFT-a N 2007 2182 1391 1864 1908 1808 2485 1710 1385 Crude model 0 (ref.) 0.023 (-0.040, 0.085) 0.072 (0.0005, 0.144)a 0.054 0 (ref.) -0.126 (-0.193, -0.059)a -0.207 (-0.272, -0.141)a <0.001 0 (ref.) -0.044 (-0.108, 0.020) -0.063 (-0.129, 0.002) 0.055 Model 1 0 (ref.) 0.042 (-0.006, 0.090) 0.066 (0.012, 0.120)a 0.015 0 (ref.) -0.020 (-0.072, 0.032) 0.005 (-0.047, 0.056) 0.870 0 (ref.) -0.007 (-0.056, 0.042) -0.032 (-0.083, 0.019) 0.222 Model 2 0 (ref.) 0.032 (-0.016, 0.079) 0.049 (-0.005, 0.103) 0.069 0 (ref.) -0.016 (-0.067, 0.035) -0.004 (-0.055, 0.047) 0.887 0 (ref.) -0.003 (-0.051, 0.045) -0.036 (-0.086, 0.014) 0.167 VFT-p N 2007 2183 1390 1864 1908 1808 2486 1709 1385 Crude model 0 (ref.) 0.047 (-0.018, 0.111) 0.055 (-0.018, 0.128) 0.119 0 (ref.) -0.083 (-0.151, -0.015)a -0.166 (-0.233, -0.098)a <0.001 0 (ref.) -0.032 (-0.098, 0.035) -0.011 (-0.078, 0.057) 0.726 Model 1 0 (ref.) 0.028 (-0.020, 0.075) 0.006 (-0.047, 0.058) 0.735 0 (ref.) -0.028 (-0.078, 0.022) -0.011 (-0.062, 0.040) 0.664 0 (ref.) -0.039 (-0.087, 0.009) 0.019 (-0.031, 0.068) 0.526 Model 2 0 (ref.) 0.020 (-0.026, 0.067) -0.001 (-0.053, 0.052) 0.928 0 (ref.) -0.025 (-0.074, 0.025) -0.02 (-0.071, 0.031) 0.432 0 (ref.) -0.030 (-0.077, 0.018) 0.015 (-0.035, 0.064) 0.612 TMT-A N 2004 2176 1390 1862 1908 1800 2485 1705 1380 Crude model 0 (ref.) 0.003 (-0.059, 0.064) -0.060 (-0.129, 0.009) 0.114 0 (ref.) 0.134 (0.076, 0.193)a 0.287 (0.221, 0.353)a <0.001 0 (ref.) 0.046 (-0.014, 0.105) 0.046 (-0.023, 0.115) 0.177 Model 1 0 (ref.) -0.029 (-0.079, 0.021) -0.079 (-0.135, -0.022)a 0.007 0 (ref.) 0.036 (-0.012, 0.083) 0.064 (0.008, 0.120)a 0.026 0 (ref.) 0.020 (-0.028, 0.069) -0.008 (-0.067, 0.050) 0.810 Model 2 0 (ref.) -0.012 (-0.062, 0.037) -0.060 (-0.117, -0.003)a 0.047 0 (ref.) 0.034 (-0.013, 0.081) 0.071 (0.014, 0.127)a 0.014 0 (ref.) 0.023 (-0.031, 0.076) -0.017 (-0.077, 0.044) 0.979 Continued on page 2.
Page 2 of 2 Supplemental Table 1. Longitudinal 2-year analysis of multivariable adjusted mean change (β [95% CI]) in cognitive function assessments according to tertiles in level of adherence to each dietary pattern at baseline (continued from page 1). Dietary Pattern MedDiet DASH MIND Neurological Assessment1 Lowest Adherence Modest Adherence Highest Adherence p for trend Lowest Adherence Modest Adherence Highest Adherence p for trend Lowest Adherence Modest Adherence Highest Adherence p for trend Score, median (range) 6 (1 to 7) 8 (8 to 9) 10 (10 to 14) 19 (8 to 21) 24 (22 to 26) 30 (27 to 38) 8 (2.5 to 8.5) 9 (9.0 to 9.5) 10.5 (10.0 to 13.5) N 1904 1790 914 1541 1578 1489 2058 1416 1134 TMT-B N 1999 2174 1384 1858 1900 1799 2477 1703 1377 Crude model 0 (ref.) -0.039 (-0.102, 0.024) -0.079 (-0.148, -0.010)a 0.026 0 (ref.) 0.128 (0.065, 0.190)a 0.285 (0.218, 0.352)a <0.001 0 (ref.) 0.082 (0.019, 0.145)a 0.067 (0.0003, 0.135)a 0.039 Model 1 0 (ref.) -0.054 (-0.104, -0.005)a -0.079 (-0.134, -0.023)a 0.004 0 (ref.) 0.032 (-0.018, 0.082) 0.045 (-0.009, 0.100) 0.102 0 (ref.) 0.051 (0.002, 0.101)a 0.015 (-0.039, 0.069) 0.527 Model 2 0 (ref.) -0.037 (-0.086, 0.011) -0.062 (-0.117, -0.007)a 0.024 0 (ref.) 0.026 (-0.022, 0.075) 0.051 (-0.003, 0.104) 0.062 0 (ref.) 0.045 (-0.003, 0.094) 0.022 (-0.031, 0.075) 0.382 DST-F N 1712 1859 1199 1642 1624 1504 2140 1456 1174 Crude model 0 (ref.) 0.066 (0.00001, 0.133) 0.072 (-0.004, 0.147) 0.048 0 (ref.) -0.073 (-0.144, -0.002)a -0.155 (-0.224, -0.086)a <0.001 0 (ref.) -0.047 (-0.114, 0.021) -0.023 (-0.097, 0.050) 0.492 Model 1 0 (ref.) 0.052 (0.001, 0.104)a 0.039 (-0.022, 0.100) 0.159 0 (ref.) -0.010 (-0.065, 0.046) 0.012 (-0.044, 0.068) 0.683 0 (ref.) -0.055 (-0.107, -0.003) -0.010 (-0.069, 0.048) 0.654 Model 2 0 (ref.) 0.049 (-0.003, 0.100) 0.037 (-0.024, 0.098) 0.184 0 (ref.) -0.001 (-0.057, 0.054) 0.016 (-0.041, 0.072) 0.584 0 (ref.) -0.043 (-0.095, 0.009) -0.007 (-0.065, 0.051) 0.758 DST-B N 1712 1858 1199 1642 1623 1504 2139 1456 1174 Crude model 0 (ref.) -0.024 (-0.089, 0.042) -0.006 (-0.082, 0.070) 0.816 0 (ref.) -0.160 (-0.230, -0.090)a -0.272 (-0.341, -0.202)a <0.001 0 (ref.) -0.034 (-0.101, 0.033) -0.014 (-0.086, 0.058) 0.673 Model 1 0 (ref.) 0.013 (-0.038, 0.064) 0.035 (-0.025, 0.095) 0.257 0 (ref.) -0.061 (-0.116, -0.006)a -0.045 (-0.102, 0.011) 0.111 0 (ref.) -0.005 (-0.057, 0.047) 0.057 (-0.001, 0.113) 0.062 Model 2 0 (ref.) 0.005 (-0.046, 0.055) 0.027 (-0.033, 0.087) 0.393 0 (ref.) -0.057 (-0.111, -0.002)a -0.049 (-0.105, 0.008) 0.089 0 (ref.) 0.006 (-0.045, 0.057) 0.055 (-0.001, 0.112)a 0.059 Model 1: Adjusted for age (in years), sex, intervention group, centre size (<250, 250 to <300, 300 to <400, ≥400), corrected for clusters (to account for couples living in the same household being randomized as a single unit), and respective cognitive test score at baseline. Model 2: Model 1 plus additional adjustment for baseline education level (primary school, secondary school, college), civil status (single, divorced or separated, married, widower), smoking habits (smoker, former smoker, never smoked), BMI (kg/m2), hypertension (yes/no), hypercholesterolemia (yes/no), diabetes (yes/no), and depressive symptomology (yes/no), baseline physical activity (MET min/week) and total energy intake (kcal/day). 1For the neurological tests, a positive value indicates better cognitive performance according to the associated test, except for TMT-A and TMT-B where a negative result indicates better performance. 2A composite of z-scores was used to calculate GCF using the formula: GCF=(ZMMSE + ZCDT + ZVFT-a + ZVFT-b + (–ZTMT-A ) + (–ZTMT-B) + ZDST-f + ZDST-b) / 8. arepresents a significant difference (p<0.05) from the reference. Abbreviations: CDT, Clock Drawing Test; DASH, Dietary Approaches to Stop Hypertension; DST-b, Digit Span test backward; DST-f, Digit Span test forward; GCF, Global Cognitive Function; MedDiet, Mediterranean dietary pattern; MIND, Mediterranean-DASH Intervention for Neurodegenerative delay; MMSE, Mini-Mental State Examination; TMT-A, Trail Making Test Part A; TMT-B, Trail Making Test Part B; VFT-a, Verbal Fluency tasks semantical; VFT-p, Verbal Fluency tasks phonological.