- Click here for explanation of classification scheme.

To determine whether regional differences persist even though the current global economic structure provides access to different food products in an unprecedented way.

Adults.

None reported.

• Recruitment: Recruitment method was not described
• Design: Cross-sectional design
• Blinding used: None
• Intervention: Not applicable
• Statistical analysis:
  • Student's T-test was used testing independent samples of omnivores and vegetarians
  • Differences among geographical regions were tested using one-way analysis of variance (with post hoc Tukey test)
  • Cluster analysis was used for analyzing questionnaire data.

• Timing of recruitment: Individuals were recruited between September 2000 and September 2004
• Dependent variables: Stable-isotope content of fingernails (δ13C and δ15N)
• Independent variables:
  • Food items available in regions (from questionnaire given to restaurants and supermarkets, as well as collection of locally producted food items). Foods were analyzed using stable-isotope ratio analysis.
  • Individual food intake (gathered from a brief questionnaire providing general details of diet)
  • Subjects were categorized as ovo-lacto-vegetarians or omnivores based on answers to the questionnaire 
• Confounding variables: Sex and age.

• Initial N: Number of individuals:
  • 273 from Southeast Brazil
  • 22 from Sanarém, Pará, Brazil (city)
  • 15 from Sao Jorge, Brazil (forest)
  • Six from Jamaraquá (river)
  • 10 from Socoro (lake)
  • 490 from the US
  • 35 from five countries in Western Europe
• Attrition (final N): Not available
• Age: Adult
• Ethnicity: Not reported
• Other relevant demographics: "Mostly" middle class
• Anthropometrics: Not reported
• Location: Brazil, US and Europe.

 

Food access is becoming increasingly global. Foods once available only in one part of the world are becoming increasingly accessible in far-flung parts of the globe. We might think that the creation of a global food system (what the authors call the “global supermarket) would increase the diversity of foods available and so decrease regional differences in foods and the physiological effects of foods.

In short, vegetarians in Brazil should resemble vegetarians in the US more and more. The same pattern should also hold for omnivores if the “global supermarket” idea is correct.

But, is this what we see?

The authors say that regional differences persist in diet even though access to foods from different parts of the globe has increased. They measure stable-isotope ratios (d15N and d13C) in individual's fingernails to determine whether region still makes a difference.

If region still makes a difference, we should see not only a difference between vegetarians and omnivores in terms of isotope content of fingernails, but we should also see a difference between regions even when people are on the same type of diet (e.g., vegetarians in the US will be different than vegetarians in Brazil).

First, Do We See Differences in Stable-Isotope Content of Fingernails?

Omnivores

The authors compared US (Salt Lake City, UT) populations with populations from South Eastern Brazil (with large urban centers) and Western Europe. They found no significant difference between males and females or differences by age.

• SE-Brazil had higher d13C and d15N values (-15.4±1.0, 10.4±0.7) than Western USA (-18.8±0.8 and 9.4±0.6) and W-Europe (-20.2±1.2 and 9.6±0.8) (P<0.05).

• US values were higher than Western European values.

So, in terms of omnivores, we do see a regional difference.

Omnivores Vs. Vegetarians

Among vegetarians, SE-Brazil was higher in d13C and d15N than individuals from Western US (P<0.05)

In Brazil

• In vegetarian/omnivore comparison, no statistical difference in value in d15N

• d13C value in vegetarians significantly lower than that of omnivores in Brazil (P<0.05).

In US

• Vegetarians had d15N values that were significantly lower (P<0.05) than omnivores in Western US

• US vegetarians and omnivores were not significantly different in their d13C values.

So, not only were vegetarians in Brazil different than vegetarians in the US (Brazilians higher for both isotopes), but the ways in which vegetarians were different than omnivores in the respective countries were different.

Could Differences in Isotope Content Be Due to Differences in Foods?

In Brazil

• Beef and sugar, followed by cereal and diary-derived products were more closely related to d13C than d15N in omnivores

• Fruits, vegetables and grains in the diet were more closely related to d15N than d13C in omnivores

• These patterns were the same in Brazilian vegetarians (excluding meat).

In the US

• Dairy-derived products, meat, fruit and vegetables more closely related to d15N than d13C in omnivores

• Cereals were more closely related to d13C values in omnivores

• These patterns were the same in US vegetarians (excluding meat).

In short, isotope content in fingernails was associated with different foods in different regions.

Why Might We See These Differences?

So, we see that not only are vegetarians and omnivores different, but that regions are different. Additionally, we see that the differences between vegetarians and omnivores are not the same in different regions.

Why?

Differences Among Food Types

The authors test for differences among food types to see whether this may explain regional differences. They classify different types of foods into four main categories:

1. C3 plants (C3 cereals—barley, oats, rice, wheat—tuber, vegetables, fruit)

2. C4 plants (C4 cereals—maize, and others)

3. Animal and products

4. Seafood

They tested to see whether there was a difference among these four categories (for instance, is there a difference between C3 foods and C4 foods?) They found that there was a significant difference in the d13C and d15N values as they relate to fingernail isotope content across categories.

Differences Within Food Groups

The authors then tested to see whether there was a difference within each category in different regions (for instance, were C3 plant foods in Brazil different than C3 plant foods in the US?).

They also found significant differences between Brazil and the US i terms of the d15N content of C3 cereals, tubers and fruits. Brazil showed heavier d15N values than the US by 0.9% to 6.2%. Also, it appeared that, overall, US values showed less variation in d15N values within these foods. In short, not only were Brazilian foods in this category higher than US foods, but there was also a wider range of these values in Brazilian foods. US C3 cereals, tubers and fruits had a more restricted range of d15N levels.

The authors also found a significant difference in dN15 values in animal products between Brazil and the US as well.

What Explains Regional Differences? Why Isn't a Tuber Just a Tuber?

The authors then go on to examine why there would be differences between plants of the same type and the same types of meats in the different regions.

• C3 plants in Brazil are higher in d15N than some of the same plants in the US. This caused the d15N levels in Brazilian vegetarians to be closer to that of US omnivores. Also, Brazilians eat more of C3 plants while individuals in the US consume more C4 plants (which helps explain the higher levels of d13C levels in the SE Brazilian population).

• But why would C3 plants in Brazil be higher than in the US? The authors suggest that part of the explanation could be differences in the use of synthetic fertilizers. Using synthetic fertilizers on plants (both for human and animal consumption) decreases the d15N content. The US uses about six times more synthetic fertilizer than Brazil.

In short, some physiological differences among vegetarians may vary by region because:

• While vegetarians in different regions exclude meat from their diet, they eat different types of plant foods (e.g., more C3 or C4 plant foods)

• The same plant foods may be different in different regions because of the plants are grown in different ways (using synthetic fertilizers or not).

With regard to meat products, some part of the variation between regions can be due to the differences in the plants fed to the animals (e.g., grains grown with synthetic fertilizers in the US). Differences in plant foods create differences in terrestrial animal foods.

Urban and Rural

Of course, regional variation does not explain everything. The authors suggest that as food accessibility becomes more global (foods that used to be available only in one part of the globe now being available across the globe) we should still see differences between areas that are more urban (and have greater access to a wider variety of foods) and those that are less integrated into the global food system. In fact, this is what they find. Comparing different regions in Brazil, the authors report that the isotope composition of fingernails in urban centers in SE Brazil was quite different than those of more isolated communities in the Amazonian region.

• Despite increased access to food because of the development of a "global supermarket," regional differences between Southeastern Brazil and the Western US persist when measured in terms of fingernail stable-isotope content. These differences appear to be associated with regional agricultural and animal production practices.
• Carbon and nitrogen isotopes of human fingernails hold dietary information directly related to both food sources and dietary practices in a region.

University/Hospital:

Centro de Energia Nuclear na Agricultura - Univ of Sao Paulo

Very detailed description of the methods of stable-isotope ratio analysis, but very weak description of sampling design.