Table 1.
Summary of the included studies.
| Study no. | Study author | Study design (Country of origin) | Age (years) | N | Food frequency assessment | MedDiet assessment | Cognitive assessment | Cognitive sub-measures | Outcome |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Scarmeas et al. (69) | Prospective cohort (USA) | 76+ | 1,984 | 61-item FFQ. Willett et al. (70) | Scored as in Trichopoulou et al. (26) | Visual Retention Test, Weschler Adult Intelligence Scale (WAIS) | Storage and retention, disordered memory and learning, visual retention | Higher adherence to a MedDiet decreases risk of AD |
| 2 | Féart et al. (71) | Prospective cohort (France) | 65+ | 1,410 | FFQ and 24-h dietary recall | Scored as in Trichopoulou et al. (26) | MMSE, Isaac Set Test, Benton Visual Retention Test, Free and Cued Selective Reminding Test | Orientation, registration, attention, calculation, recall and language, controlled attention memory, semi-verbal fluency, speed of verbal production | MedDiet slower MMSE cognitive decline and not risk for incident dementia |
| 3 | Scarmeas et al. (72) | Multiethnic community longitudinal study (USA) | 76–78 | 1,393 | FFQ. Willett (70) | Scored as in Trichopoulou et al. (26) | Neuropsychological battery, Global Clinical Dementia Rating (CDR) | Long- and short-term memory, judgment, and problem-solving | Higher adherence to a MedDiet associated with reduced risk of MCI, reduced conversion of MCI to AD |
| 4 | Gu et al. (73) | Prospective cohort (USA) | 65+ | 1,219 | Semi-qualitative 61-item FFQ. Willett (70) | Scored as in Scarmeas et al. (72) and Trichopoulou et al. (26) | 15 neuropsychological tests | Long- and short-term verbal memory, short-term non-verbal memory, orientation, construction, abstract reasoning, language | Higher adherence to a MedDiet associated with reduced risk of AD |
| 5 | McMillan et al. (74) | Randomized single-blind parallel groups (Australia) | 19–30 | 27 | FD | None, MedDiet intervention with 93% adherence | COMPASS | Attention, working memory, long-term working memory, executive function, response time, word recognition | MedDiet increases vigor, alertness, contentment, changes in cognitive function |
| 6 | Tangney et al. (34) | Data analysis of the CHAP study Longitudinal study (USA) |
≥65 | 3,790 | Harvard FFQ | Scored as in Panagiotakos et al. (75) | MMSE, East Boston tests of immediate and delayed recall, symbols digit modalities test | Immediate and delayed recall, orientation, registration, attention, calculation, recall and language | MedDiet associated with reduced rate of cognitive decline with older age |
| 7 | Cherbuin and Anstey (35) | Longitudinal study (Australia) | 60–64 | 1,528 | CSIRO FFQ | Scored as in Scarmeas (72) and Trichopoulou (26) | International consensus criteria, CDR | Perceptual memory, motor function, simple and complex reaction time | MedDiet not protective against cognitive decline |
| 8 | Vercambre et al. (76) | Prospective cohort (women only) (USA) | 65+ | 2,504 | FFQ 116 food groups. Willet (70) | Scored as in Trichopoulou (26) | TICS, telephone adaptation of the MMSE, verbal memory with the TICS 10-word list, East Boston memory test (EBMT) | Immediate and delayed recall, orientation, registration, attention, calculation, recall, and language | No association between MedDiet and 5-year cognitive change |
| 9 | Kesse-Guyot et al. (62) | Prospective cohort (France) | 65.4 ± 4.6 | 3,083 | French food composition table, computerized questionnaire FFQ | Mediterranean Dietary Pattern Score. Rumawas et al. (77) | Rappel indices 48-item (cued recall), verbal fluency tests of lexical/semantic memory | Episodic memory, working memory, verbal fluency, phonometric fluency | MedDiet adherence was not associated with cognitive performance overall |
| 10 | Martinez-Lapiscina et al. (78) | Multicenter randomized (Spain) | 74.6 ± 5.7 | 522 | Semi-quantitative FFQ. Fernández-Ballarth et al. (79) | 14-item questionnaire. Martínez-gonzález et al. (27) | MMSE and Clock-Drawing Test (CDT), CDT language and comprehension | Working memory, visual and spatial orientation, orientation, registration, attention, calculation, recall and language | Higher adherence to a MedDiet intervention improved cognition compared with a low-fat diet |
| 11 | Samieri et al. (80) | Prospective epidemiological (USA) | ≥70 | 16,058 | USDA FFQ | Scored as in Trichopoulou (26) | TICS, immediate and delayed recall of the EBMT, delayed recall of the TICS 10-word list, category fluency, Digital Spin backward | Global cognitive scores, immediate and delayed recall, verbal memory, language | Long-term adherence to the MedDiet pattern is modestly associated with global cognitive function and verbal memory in later life, but not with a cognitive change after a 6-year period |
| 12 | Samieri et al. (81) | Randomized double-blind (USA) | 65+ | 6,174 | 131-item semi-quantitative FFQ. Rimm et al. (82) | Scored as in Fung et al. (83) | TICS, immediate and delayed recall of the EBMT, delayed recall of the TICS 10-word test, category fluency | Recall memory, working memory, immediate and delayed recall, language | MedDiet not associated with cognitive decline; though certain dietary components may be related and warrant further investigation |
| 13 | Titova et al. (84) | Prospective cohort (Sweden) | 70 | 194 | FFQ. Rimm (82) | Scored as in Trichopoulou (26) | Swedish 7-min screen for cognitive decline and dementia, Benton temporal orientation, CDT, categorical verbal fluency | Recall memory, temporal orientation, attention, verbal fluency | Reduced intake of meat and meat products associated with greater brain volume and better cognitive performance |
| 14 | Tsivgoulis et al. (85) | Prospective cohort (USA) | 45–90 | 17,478 | Food intake via a 1-week diary | Scored as in Scarmeas et al. (72) and Féart et al. (71) | 6-item screen. Callahan et al. (86) | Delayed memory | High adherence to a MedDiet associated with reduced rate of incident cognitive impairment |
| 15 | Ye et al. (87) | Longitudinal cohort (Puerto Rican adults in USA) | 45–75 | 1,269 | NHNS FFQ | Scored as in Trichopoulou (26) | MMSE, word learning list forward and back, digit spin Stroop test, verbal fluency | Memory, attention, orientation, registration, calculation, recall and language learning, immediate recall, recognition and percent recognition | High adherence to a MedDiet associated with greater cognitive function and reduced risk of cognitive impairment |
| 16 | Galbete et al. (88) | Prospective cohort (Spain) | 62 ± 6 | 824 | Validated semi-quantitative FFQ, MyPyramis Equivalent Database 2.0 for USDA Survey Foods 2003–2004. Bowman et al. (89) | Scored as in Trichopoulou (26) | Spanish TICS | Immediate memory, delayed recall, orientation, attention, calculation, language | Higher adherence to a MedDiet associated with better cognitive outcomes |
| 17 | Lee et al. (90) | Randomized trial – balanced crossover groups (Australia) | 20–38 | 24 | FD | None, ≥80% adherence for dietary intervention | COMPASS | Attention, working memory, long-term memory, executive function | A MedDiet potential to enhance mood, cognition, and cardiovascular health in young adults |
| 18 | Valls-Pedret et al. (91) | Randomized trial – parallel groups (Spain) | 67 (mean) | 447 | MedDiet program | None/high compliance | MMSE, Rey Auditory Verbal Learning Test, Animals Semantic Fluency, Digit Span from the WAIS, Verbal Paired Association | Memory, attention, executive function | A MedDiet is associated with improved cognitive function |
AD, Alzheimer’s disease; FD, food diary; FFQ, Food Frequency Questionnaire; CDR, global clinical dementia rating; CDT, clock-drawing test; CHAP, Chicago Health and Aging Project; COMPASS, computerized mental performance assessment system; CSIRO, Commonwealth Scientific and Industrial Research Organisation; EBMT, East Boston memory test; MedDiet, Mediterranean diet; MCI, mild cognitive impairment; MMSE, mini–mental state examination; TICS, telephone interview for cognitive status; USA, United States of America; USDA, United States Department of Agriculture; WAIS, Wechsler Adult Intelligence Scale.