Skip to main content
. 2017 Apr 27;12(4):e0175149. doi: 10.1371/journal.pone.0175149

Table 4. Data sources and identified optimal intake levels of specific dietary factors for reducing cardiometabolic diseases1.

Dietary Factor (standardized serving size) Observed intake levels associated with lowest disease risk in meta-analyses (health outcome) Observed mean national intakes in 20102 Recommended intakes by major dietary guidelines3 Optimal mean population intake4
FOODS
Fruits5 (100 g/serving)

  • 2.4 servings/d (CHD)

  • 3.4 servings/d (ischemic stroke)

  • 2.1 servings/d (hemorrhagic stroke)

 Top 3 countries:

  • Barbados: 410 g/d (4.1 servings/d)

  • Jamaica: 315 g/d (3.2 servings/d)

  • Malaysia: 299 g/d (3.0 servings/d)

 DGA 2015: 2 cups/d

  • AHA 2020: ≥4.5 cups/d (fruits and vegetables)

 3 (100 g) servings/d
Vegetables (including beans/legumes)6 (100 g/serving)

  • 3.7 servings/d (CHD)

  • 3.4 servings/d (ischemic stroke)

  • 1.5 servings/d (hemorrhagic stroke)

 Top 3 countries:

  • Lao PDR: 364 g/d (3.6 servings/d)

  • Bhutan: 302 g/d (3.0 servings/d)

  • Taiwan: 293 g/d (2.9 servings/d)

  • DGA 2015: 1.8 cups/d (excluding starchy vegetables)

  • AHA 2020: ≥4.5 cups/d (fruits and vegetables)

 4 (100 g) servings/d
Beans/legumes (100 g/serving)

  • 4.2 servings/wk (CHD) (0.6 servings/d)

 Top 3 countries:

  • Brazil: 194 g/d (1.9 servings/d)

  • Colombia: 137 g/d (1.4 servings/d)

  • Mexico: 103 g/d (1.0 serving/d)

  • DGA 2015: 1 ½ cups/wk (0.2 servings/d)

 1 (100 g) serving/d
Nuts/seeds (1 oz [28.35 g]/serving)

  • 5.0 servings/wk (fatal CHD)

  • 5.2 servings/wk (non-fatal CHD)

  • 4.9 servings/wk (diabetes)

 Top 3 countries:

  • Malaysia: 74 g/d (18.3 servings/wk)

  • Lebanon: 24 g/d (6.0 servings/wk)

  • UK: 15 g/d (3.7 servings/wk)

  • DGA 2015: 5 oz/wk (including soy products)

  • AHA 2020: ≥4 servings/wk7

 5 (1 oz) servings/wk
Whole grains (50 g/serving)

  • 2.5 servings/d (CVD)

  • 2.5 servings/d (CHD)

  • 3.0 servings/d (diabetes)

 Top 3 countries:

  • Germany: 128 g/d (2.6 servings/d)

  • Barbados: 118 g/d (2.4 servings/d)

  • Australia: 93 g/d (1.9 servings/d)

 DGA 2015: ≥3 (1 oz) servings/d (≥1.7 servings/d)

  • AHA 2020: ≥3 (1 oz) servings/d (≥1.7 servings/d)

 2.5 (50 g) servings/d
Red meats, unprocessed (100 g/serving)

  • 0.19 servings/d (diabetes) (1.3 servings/wk)

 Bottom 3 countries:

  • Indonesia: 12 g/d (0.8 servings/wk)

  • Armenia: 15 g/d (1.0 servings/wk)

  • Georgia: 15 g/d (1.0 servings/wk)

  • DGA 2015: 26 oz/wk (lean meat, poultry and eggs) (7.4 servings/wk)

 1 (100 g) serving/wk
Processed meats (50 g/serving)

  • 0.07 serving/d (CHD) (0.5 servings/wk)

  • 0.11 serving/d (diabetes) (0.8 servings/wk)

 Bottom 3 countries:

  • Iran: 2.5 g/d (0.4 servings/wk)

  • Korea: 3.1 g/d (0.4 servings/wk)

  • China: 3.3 g/d (0.5 servings/wk)

  • DGA 2015: Choose fresh lean meat, rather than processed meat

  • AHA 2020: none or ≤2 servings/wk

 0
Fish/Seafood (100 g/serving)

  • 3 servings/wk (fatal CHD)

  • 6.5 servings/wk (total stroke)

 Top 3 countries:

  • Japan: 105 g/d (7.4 servings/wk)

  • Korea: 73 g/d (5.1 servings/wk)

  • Iceland: 66 g/d (4.6 servings/wk)

  • DGA 2015: 8 oz/wk (2 servings/wk)

  • AHA 2020: ≥2 (3.5 oz) servings/wk (preferably oily fish) (≥2 servings/wk)

 3.5 (100 g) servings/wk
Yoghurt (8 oz (244 g)/serving)

  • 2.6 servings/wk (diabetes)

 Not available

  • DGA 2015: 3 cups/d of dairy products (21 cups/wk)

 2.5 (8 oz) servings/wk
Sugar-sweetened beverages (8 oz (236.5 g)/serving)

  • 0 servings/d (body mass index)

  • 0.017 serving/d (diabetes)

 Bottom 3 countries:

  • China: 14 g/d (0.06 servings/d)

  • Iran: 24 g/d (0.1 servings/d)

  • Poland: 24 g/d (0.1 servings/d)

  • DGA 2015: <10%E from added sugars

  • AHA 2020: ≤450 kcal (36 oz)/wk (≤0.6 servings/d)

 0
NUTRIENTS
Polyunsaturated fats replacing carbohydrates or saturated fats (%E)

  • 5.1%E (CHD, cohorts)

  • 14.9%E (CHD, randomized trials)

 Top 3 countries, PUFA:

  • Bulgaria: 11.3%E

  • Lebanon: 9.3%E

  • Hungary: 8.9%E

Bottom 3 countries, SFA:

  • Japan: 7.0%E

  • Mexico: 7.4%E

  • Barbados: 7.6%E

  • DGA 2015: <10%E from saturated fat replaced with unsaturated especially polyunsaturated fats

  • FAO 2010: 6–11%E from PUFA

 11%E 9
Seafood omega-3 fats (mg/d)

  • 250 mg/d (fatal CHD)

 Top 3 countries:

  • Barbados: 1,191 mg/d

  • Iceland: 1,120 mg/d

  • Japan: 993 mg/d

  • DGA 2015: 250 mg/d8

 250 mg/d
Trans-fats (%E)

  • 0%E (CHD)

 Bottom 3 countries:

  • Barbados: 0.2%E

  • Finland: 0.4%E

  • Italy: 0.5%E

 DGA 2015: As low as possible

  • WHO: <1%E [97]

 0.5%E10
Dietary fiber (g/d)

  • 23.8 g/d (CVD)

  • 22.8 g/d (CHD)

  • 19.4 g/d (stroke)

  • 30.0 g/d (diabetes)

 Top 3 countries:

  • Barbados: 28 g/d

  • Mexico: 26 g/d

  • Bulgaria: 25 g/d

  • DGA 2015: 22.4–33.6 g/d (depending on gender & age group)

 30 g/d
Sodium (mg/d) 614 mg/d (lower SBP)11 [60]

  • 1,500 mg/d (reduced BP in randomized trials) [46]

  • 1,787 mg/d (CHD mortality) [61]

  • 2,245 mg/d (stroke mortality) [61]

  • 2,391 mg/d (stroke)

 5 nations with mean intakes at or below 2,000 mg/d: Jamaica, Colombia, South Africa, Mexico, Iran

  • UK NICE: <1,200 mg/d [62]

  • AHA 2020: <1,500 mg/d [63]

  • WHO: 2,000 mg/d [64]

  • DGA 2015: <2,300 mg/d

  • UK FSA: <2,400 mg/d [65]

 2,000 mg/d12
Potassium (mg/d)

  • 4,136 mg/d (stroke)

 Not available

  • DGA 2015: 4,700 mg/d

  • WHO: ≥3,500 mg/d [98]

 4,500 mg/d

1 Building up on our work in the 2010 Nutrition and Chronic Diseases Expert Group (NutriCoDE) in which we evaluated optimal intake levels for cardiometabolic outcomes including coronary heart disease (CHD), stroke, type 2 diabetes, body mass index (BMI), and systolic blood pressure (SBP) [10]. For both studies of clinical outcomes and national surveys of dietary intakes, we used standardized servings to account for any variability in serving sizes across studies or countries. Thus, the characterized RR’s are accurate for the listed serving size. For populations with smaller or larger serving sizes, the RR’s should be appropriately adjusted.

2 Based on nationally representative, individual-level dietary surveys using optimal dietary metrics among both adult men and women in our 2010 NutriCoDE Global Dietary Database [27–31] and other sources [32]; adjusted to 2000 kcal/d.

3 For an average intake of 2,000 kcal/d.

4 For an average energy intake of 2,000 kcal/d. The optimal mean levels for the population were determined based on risk (observed levels at which lowest disease risk occurs), feasibility (observed national consumption levels globally), and consistency (with other assessments in major dietary guidelines) [33, 34]. The plausible population distribution of consumption (SD) around the optimal population mean was determined to be ±10% of the mean, based on the average SD for diet-related metabolic risk factors [3, 36–39]. We could not comparably identify optimal intake levels of glycemic load due to absence of global data on mean intakes in most nations and of recommended levels in major dietary guidelines.

5 Excluding 100% juices.

6 Excluding vegetable juices, starchy vegetables such as potatoes or corn, and salted or pickled vegetables.

7 Including beans/legumes.

8 Based on eicosapentaenoic acid (EPA) + docosahexaenoic acid (DHA) in common fish varieties, the DGA 2015 calculates that 1,750 mg/wk (250 mg/d) would be in concordance with the recommended fish intake of 8 oz/wk.

9 12%E in earlier analyses [3, 25, 29]. Lowered to 11% based on the present updated review of available evidence on optimal levels and considering observed national intakes and major dietary guidelines.

10 Non-zero value to account for natural ruminant sources, for which probable or convincing evidence of causal effects on cardiometabolic outcomes was not identified.

11 Based on ecologic evidence; these values are the mean 24-hour urine sodium excretion across the 4 populations with lowest levels (Brazil, Yanomano and Xingu, Papua New Guinea, Kenya) in the Intersalt Study [60].

12 As previously described in detail, we did not incorporate a potential U-shaped relationship with risk due to the linear dose-response effect of dietary sodium on BP, the log-linear effect of BP on cardiovascular disease (CVD), the absence of plausible biologic rationale for increased risk with sodium reduction, at least to 2,000 mg/d, and the plausible source of bias that could explain the U-shaped relationships observed in some, but not all, prior observational studies [24].

AHA 2020, American Heart Association 2020 Strategic Impact Goals [35]; DGA, Dietary Guidelines for Americans [33]; FAO, United Nations Food and Agricultural Organization [34]; PUFA, Polyunsaturated fats; SFA, Saturated fats; UK FSA, Food Standards Agency; UK NICE, UK National Institute for Health and Clinical Excellence; WHO, World Health Organization.