The only reason I picked up on the change was because the figure in the original publication (reference below) had such a striking problem when it was published.
The analytic method used caused sugar sweetened beverages (SSBs) and other highly processed foods to be presented as “healthful” relative to unprocessed foods.
What this shows is that the analysis puts emphasis on foods that are artificial and cheap to produce (low environmental impact) without regard to nutrient density or metabolic health impact. We now know that the benefits of easy-to-produce engineered foods with stable shelf lives and other characteristics come at an enormous cost in population health. Taking “SSBs” out removed that red flag and covers up a significant imbalance noted in the original findings.
When The Economist remixed the original, this cognitive dissonance was removed, making the interpretation much more consistent with their headline. Additional changes were made, including the scale of the axes (from log to linear, shifting the x axis), which creates greater visual space between “good” and “bad.” This also feeds the headline of the article.
There are other critiques of the source of the data in the scientific article which have been covered elsewhere, so I won’t do that here, however, if there are questions about these points, I am happy to answer them.
- Relative risk magnifies perceived harm and is suboptimal for communication compared to absolute risk
- The data comes from epidemiology studies, not clinical trials
- It’s not clear if referenced studies controlled for healthy user bias – the paper says “age, body mass index, gender, and smoking” only
- Olive oil is mentioned as a proxy for all vegetable oils, including highly refined and ultra-processed oils, which have very different characteristics and health effects
Dietary impact on GHG emissions is not discussed in the context of much greater discretionary sources
I recently spent some time investigating this for myself, because healthy people, healthy planet is my interest, not one or the other.
Based on the headline of The Economist piece, the answer may be “not very much,” and here’s why:
Given that fossil fuel use has far greater impacts on the environment, I have personally decided to add renewable energy to the global electricity grid. This is in addition to not owning a car, of course 🙂 .
Americans are getting most of their nutrition from plants already
…and unfortunatey, they have been choosing the most unhealthy kinds of plants.
The Economist didn’t include this information, or information about the impacts on the environment from increased health care use arising from these choices (see: Just Read: Carbon footprint of the global pharmaceutical industry – “Significantly worse than the automotive industry” ).
1. Clark MA, Springmann M, Hill J, Tilman D. Multiple health and environmental impacts of foods. Proc Natl Acad Sci [Internet]. 2019 Oct 28;201906908. Available from: http://www.pnas.org/lookup/doi/10.1073/pnas.1906908116
2. How much would giving up meat help the environment? – Daily chart [Internet]. The Economist. 2019 [cited 2019 Nov 17]. Available from: https://www.economist.com/graphic-detail/2019/11/15/how-much-would-giving-up-meat-help-the-environment