Like many in the United States, I often heard the adage of the “starving kids in China”. Nowadays, this is no longer a phrase I hear often. It may be because I am no longer a child. The fact that not much of China is as poor or starving anymore, I think, also plays a role in this. After all China has amazingly reduced its poverty rate from 88% to just 6% in the past 3 decades.
Even starving children in Africa are now considered rare(they do suffer from malnutrition however).
We no longer need to think of starving children in other countries as a push to finish our plates and we could probably benefit from putting our fork down as well. Obesity is becoming a real issue in the United States as 1 in 3 children ages 6 to 19 are overweight or obese, with the rate for adults being twice that.
It sounds bad. It is bad. However in relation to the rest of the world, it shouldn’t be come as a shock and we are not unique as indicated by an article published by The Atlantic.
With all of this in mind I wanted to get a visual representation of this.
Results by region
As a quick reference the BMI classifications are:
| Classification | Range | Color |
|---|---|---|
| Underweight | < 18.5 | dark purple-purple |
| Normal weight | 18.5–24.9 | blue-green |
| Overweight | 25–29.9 | orange-red |
| Obesity | > 30 | red-black |
I used BMI data published by the WHO, and regional and income classifications published by the World Bank to generate the graphs. The graphs use a colored scale to indicate in what BMI range each country falls into.
As a note the WHO provides their own regional classifications for countries, but I decided to not use it as I personally found it confusing and suspected it would be for others as well. For example countries labelled as being in the WHO’s South-East Asia region include Maldives, Bhutan, Sri Lanka, Nepal, India, Bangladesh, and North Korea.
Using the World Bank’s regional grouping:
Looking at this, we can generally see that outside of the Sub-Saharan Africa, South Asia, and East Asian & Pacific, regions, the countries of the world all have an average BMI in the overweight or obese range.
Looking at the extreme ends of the BMI range, we can see that there are no countries possessing an underweight average BMI.
there are no countries possessing an underweight average BMI
At the same time, we can see that obesity has hit particularly hard to certain countries in the East Asia & Pacific region. Looking more closely, the impact of obesity has hit hardest in the Pacific Islands. Using the reasonable assumption of a normal distribution, this means at least half of all men and women in the countries of Nauru, Tonga, Samoa, and Palau are obese.
at least half of all men and women in the countries of Nauru, Tonga, Samoa, and Palau are obese
The only other country where this statement holds true is Kuwait, although there are many other countries that are just shy of making the cut. For example, here in my home country the U.S of A, male and female are only about 1 BMI point away.
Results by income
Because there is a relationship between obesity and socioeconomics, I also wanted to see if there was a visual pattern between a country’s income and its average BMI.
For reference the World Bank’s income grouping:
| Classification(2014) | GNI / capita |
|---|---|
| Low income | <= $1045 |
| Lower middle income | $1,046-$4,125 |
| Upper middle income | $4,126-$12,735 |
| Upper income | > $12,735 |
Through grouping countries by their income, we see a clear correlation between a country’s income and its average BMI.
we see a clear correlation between a country’s income and its average BMI
In the low income group, all countries have a normal average BMI except for Sudan which is disqualified by only 0.2 points.
The lower middle income group has the greatest spread with about half being in the normal range and half in the overweight or obese range. There needs to be further research into this but I speculate that being above a certain income allows countries to afford access to more “Westernized diets”, which has been speculated as a potential contribution to weight gain. This could be due to being able to afford more industrial processing for food or perhaps just greater importation. However this is all unfounded and requires further exploration.
In the upper middle and high income group, we once again see a nearly homogenous pattern. Excluding what can be labelled as Asian countries and Angola, all of the upper middle and high income countries possess an overweight or obese average BMI.
Excluding what can be labelled as Asian countries and Angola, all of the upper middle and high income countries possess an overweight or obese average BMI
A potentially interesting pattern I noticed were the countries where the average BMI for male and female was split across normal and overweight. The average female BMI was lower than the average male BMI in high income countries, but this trend was reversed in upper middle income countries. Looking into this peculiarity of gender differences actually led to me reading and writing quite a bit more than anticipated, and will be in a second part as this post is already getting a bit long.
Before ending this post, here is the BMI plotted on a world map:
Unfortunately much of the Pacific Island countries do not show up on the map perhaps due to their diminutive size but we can can still see the rest of the world.
[Part 2 coming soon]
Notes
- The data used is from 2014
- The direct link for the WHO dataset is broken. I worked around this by taking the dataset used in their Interactive Graph
- I’ve sent an email to the WHO about the broken dataset but have yet to receive a response