Table 3.
Meta-analyses for cognitive outcomes in evaluation of nutritional interventions during pregnancy.
| Dietary Nutrient | Cognitive Outcome | Study Authors | Country (OECD) | Studies (n) | Children (n) | SMD (95% CI) † | p-Value | I2 ‡ |
|---|---|---|---|---|---|---|---|---|
| LCPUFA | Attention | Gould et al. [47] Ramakrishnan et al. [74] |
Australia (1) Mexico (2) |
2 | 955 | −0.07 (−0.17 to 0.03) | 0.19 | 0% |
| LCPUFA | Behaviour | Dunstan et al. [46] Makrides et al. [52] Brei et al. [72] Ramakrishnan et al. [74] |
Australia (1) Australia (1) Germany (1) Mexico (2) |
4 | 1725 | −0.05 (−0.12 to 0.03) | 0.25 | 0% |
| MMN | Motor skills | Li et al. [60] McGrath et al. [61] Prado et al. [53] Chang et al. [73] Christian [71] Christian [67] |
China (2) Tanzania (4) Indonesia (3) China (3) Nepal (4) Bangladesh (4) |
6 | 3572 | 0.02 (−0.04 to 0.17) | 0.55 | 0% |
| LCPUFA | Motor skills | Dunstan et al. [46] Makrides et al. [52] Mulder et al. [62] Tofail et al. [63] Van Goor et al. [56] Brei et al. [72] Ramakrishnan et al. [74] |
Australia (1) Australia (1) Canada (1) Bangladesh (4) Netherlands (1) Germany (1) Mexico (2) |
7 | 2265 | 0.06 (−0.03 to 0.15) | 0.22 | 8.9% |
| Zinc | Motor skills | Caulfield et al. [57] Hamadani et al. [58] Tamura et al. [55] |
Peru (2) Bangladesh (4) United States (1) |
3 | 985 | −0.10 (−0.38 to 0.17) | 0.49 | 72.5% |
| LCPUFA | Fluid intelligence | Brei et al. [72] Dunstan et al. [46] Ramakrishnan et al. [74] |
Germany (1) Australia (1) Mexico (2) |
3 | 999 | 0.05 (−0.08 to 0.18) | 0.45 | 10.1% |
| MMN | Fluid intelligence | Christian et al. [71] Prado et al. [53] |
Nepal (4) Indonesia (3) |
2 | 755 | 0.07 (−0.20 to 0.33) | 0.63 | 78.2% |
| Zinc | Fluid intelligence | Caulifield et al. [57] Tamura et al. [55] |
Peru (2) United States (1) |
2 | 539 | −0.10 (−0.25 to 0.06) | 0.23 | 0% |
| MMN | Global cognition | Joos et al. [116] Li et al. [60] McGrath et al. [61] Waber et al. [125] Chang et al. [73] Christian et al. [67] |
Taiwan (2) China (2) Tanzania (4) Colombia (2) China (3) Bangladesh (4) |
6 | 3126 | 0.09 (−0.02 to 0.19) | 0.11 | 57.2% |
| LCPUFA | Global cognition | Dunstan et al. [46] Helland et al. [49] Helland et al. [51] Judge et al. [59] Makrides et al. [52], Mulder et al. [62] Tofail et al. [63] Van Goor et al. [56] Brei et al. [72] Ramakrishnan et al. [74] |
Australia (1) Norway (1) Norway (1) United States (1) Australia (1) Canada (1) Bangladesh (4) Netherlands (1) Germany (1) Mexico (2) |
10 | 2632 | 0.03 (−0.07 to 0.13) | 0.55 | 21.3% |
| Zinc | Crystallised intelligence | Caulfield et al. [57] Tamura et al. [55] |
Peru (2) United States (1) |
2 | 539 | −0.04 (−0.20 to 0.12) | 0.61 | 0% |
| LCPUFA | Crystallised intelligence | Dunstan et al. [46] Makrides et al. [52] Mulder et al. [62] Brie et al. [72] Ramakrishnan et al. [74] |
Australia (1) Australia (1) Canada (1) Germany (1) Mexico (2) |
5 | 1941 | 0.25 (−0.04 to 0.53) | 0.09 | 87.8% |
| MMN | Crystallised intelligence | Christian et al. [67] Prado et al. [53] |
Bangladesh (4) Indonesia (3) |
2 | 1207 | 0.01 (−0.11 to 0.12) | 0.91 | 0% |
CI: confidence interval; LCPUFA: long chain polyunsaturated fatty acids; MMN: multiple micronutrient; SMD: standardised mean difference; OECD: the organisation for economic co-operation and development criteria; 1 = high income country, 2 = higher middle income country, 3 = lower middle income country & 4 = low income country; † the main measure of effect was SMD (also known as Cohens d). The SMD was determined by taking the difference between the mean of the intervention group compared to the control group, and dividing the pooled standard deviation for the outcome across the whole trial. A random effects model using the method DerSimonian & Laird [43] was applied to the data; ‡ the I2 statistic is the percentage of observed total variation across studies that is due to heterogeneity rather than chance. It is calculated using the following formula: I2 = 100% × (Q − df)/Q, where Q is Cochran’s heterogeneity and df is the degrees of freedom [44].