Table 1.
Evidence from cohort studies according to year of publication.
| Author | Population | Measurement of Cognition (Cutoff Point) and Diet | Protein | Protein Food Source | Thiamine | Adjustments | NHMRC Level of Evidence |
|---|---|---|---|---|---|---|---|
| La Rue et al. 1997 [27], USA | 304 community-dwelling healthy individuals, age between 66 and 90 years (6 years cohort). | The Abstraction Scale from Shipley-Hartford Intelligence Test, The Logical Memory and Visual Reproduction subtests from Wechsler Memory Scale and Rey-Osterrieth Test 3-day Food Dietary records | Protein (g/day) Baseline (median): 75 g/day 6 Years Follow Up (median): 72 g/day Rey Osterrieth Recall Test r = 0.19 p-value < 0.05 Logical Memory Test r = 0.20 p-value < 0.05 Significantly positive association between dietary protein intake and Rey Osterrieth Recall and Logical Memory Test scores. |
Thiamine (mg/day) Baseline (median): 1.93 6 years Follow up (median): 2.47 Rey Osterrieth Copy Test Baseline, r = −0.07 6 years Follow up, r = 0.16 p-value < 0.10 Rey Osterrieth Recall Test Baseline, r = 0.08 6 years Follow up, r = 0.15 p-value < 0.10 Shipley Hartford Abstraction Test Baseline, r = −0.08 6 years Follow up, r = 0.29 p-value < 0.01 Significant positive association between dietary thiamine intake and Shipley Hartford Abstraction Test scores. |
Body weight | III-2, Neutral | |
| Deschamps et al. 2002 [28], France | 125 community-dwelling non-demented elderly, age >68 years (5 years cohort) 3-day food diary, Diet history and FFQ | MMSE (cognitive decline: reduction of ≥3 MMSE score over 5 years) | Protein Median intake: 1.33 g/kg/day <1.0 g/kg/day (n = 21), OR = 1.00 ≥1.0 g/kg/day (n = 104), OR = 1.92 (0.38–9.62) p-value n.s No significant association between dietary protein intake and cognitive decline. |
Age, Gender and Education | III-2, Neutral | ||
| McNeill et al. 2011 [29], UK | 882 subjects living around Edinburg, mean age 70 years. (57 to 60 years cohort) | MHT, MMSE, NART, Verbal Fluency test, Wescler Adult Intelligence Scale–III. Semi-quantitative FFQ | Thiamine (mg/day) Mean Intake from diet: 1.51 ± 0.50 MMSE β = 0.057 p-value n.s Verbal fluency β = 0.004 p-value n.s No significant association between dietary thiamine intake and cognitive function or verbal fluency. |
Age, gender, IQ at age 11 years, smoking, social class, education, statin use, presence of APOE allele | III-3, Neutral | ||
| Barberger-Gateau et al. 2007 [24], France | 8085 free-living non-demented elderly, age >65 years (4 years cohort) | DSM- IV by neurologist FFQ | Meat All cause dementia 2–3 times/week: Incidence = 1.13 (0.88–1.97), 4–6 times/week: Incidence = 0.85 (0.68–1.01), Daily: Incidence =1.03 (0.80–1.27) p-value n.s AD 2–3 times/week: Incidence = 0.76 (0.58–0.96) 4–6 times/week: Incidence = 0.53 (0.40–0.66), Daily: Incidence =0.65 (0.46–0.84) p-value n.s No significant association between meat consumption and dementia/AD. |
Age | III-2 | ||
| Velho et al. 2008 [30], Portugal | 187 free-living elderly participants with normal cognition, age >65 years (8.5 ± 3.5 months cohort). | MMSE (Improvement: Increase >0 in MMSE score, No Improvement: No increase in MMSE score) 3-day food diary | Protein (g/day) No improvement: 70.9 ± 2.0 Improvement: 73.4 ± 1.8 t-test = 1.04 p-value n.s No significant association between dietary protein intake and improvements in cognitive function. |
Age, Total energy | III-2, Neutral | ||
| Vercambre et al. 2009 [31], France | 4809 elderly women from E3N cohort, age between 63 and 68 years (13 years cohort). | DECO scale (<33), Questionnaire for close relative/friend 208 item FFQ, 24 h recalls | Protein (g/day) Mean = 87.70 ± 24.55 Q3-Q1 OR = 0.92 (0.74–1.14) Trend (p-value n.s) No significant association between dietary protein intake and cognitive function. |
Beef, pork, lamb (g/day) Mean = 45.45 ± 35.53 Q3-Q1 OR = 0.87 (0.66–1.15) Trend (p-value n.s) Poultry (g/day) Mean = 16.93 ± 17.89 Q3-Q1 OR = 0.73 (0.58–0.91) Trend (p-value = 0.004) No significant association for beef, pork and lamb. Significantly higher poultry intake in participants with better cognition. |
Age, Education, BMI, Frequency of average physical activity, Average daily energy intake, Smoking, Supplements, Post-menopausal hormones, Depression, Cancer, CHD, Stroke, T2DM, High cholesterol and Hypertension. | III-2, Neutral | |
| Roberts et al. 2012 [32], USA | 937 cognitively normal participants, age between 70 and 89 (median of 3.7 years cohort) | CDR, Short test of mental status with 9 test assessing 4 domains of memory, executive function , language and visuospatial skills 128 item FFQ | Protein (g/day) All participants (mean): 78 g/day, 18% energy Q1 (20% energy) HR = 0.79 (0.52–1.20) Correlation of trend across quartiles, p-value = 0.03 Significant association between dietary protein intake of 16%–20% of energy intake and reduced risk of MCI or dementia. |
Gender, Education, Total daily energy, Non-participation at baseline, Single macronutrient, APOE e4, T2DM, Depression, BMI, Stroke, Marital status, Smoking, Alcohol, Occupation and Frequency of moderate physical activity | III-2, Neutral |
Abbrevations: AD, Alzheimer’s Disease; CDR, Clinical Dementia Rating; DECO, Deterioration Cognitive Observée; DSM, Diagnostic and Statistical Manual of Mental Disorders; FFQ, Food Frequency Questionnaire; MHT, Moray House Test; MMSE, Mini Mental State Examination; NART, National Adult Reading Test.