GM maize data mountain

June 2018

A recent article in Forbes Magazine suggested that the "mountains of scientific proof" of "how big the benefit and small the risk is from GMO crops" should "put a categorical end to (the) worries" of the apparently uninformed public in whom "GMO paranoia continues to rage".

Five GM 'worries' were identified in the article:
  1. adverse health effects 
  2. lack of safety testing relevant to humans 
  3. environmental effects of increased pesticides 
  4. reduced crop diversity 
  5. effects on non-targeted wildlife in the GMO field. 
The 'mountain' of science referred to in the article was compiled using the internet by Italian researchers. It consists of 6,006 published articles and reviews on GM maize during the period 1996-2016, mostly examining 'Bt' insecticidal crops.

From this data base, studies from controlled field trials* were extracted and fed into a new study. This reduced the 'mountain' to a mole-hill of 76 papers (little more than 1% of the total).

* Controlled studies use genetically close, non-GM comparator crops grown side-by-side

From these 76 papers, the researchers extracted data on four agronomic, toxicological and environmental topics:
  • grain yield and quality 
  • effectiveness on the target insect 
  • effects on non-target wildlife 
  • eases of decomposition of the non-harvested crop matter in the field (i.e. nutrient cycling) 
After pooling the data on each topic, a statistical 'meta-analysis' was carried out to determine the "Impact of genetically engineered maize on agronomic, environmental and toxicological traits". Its conclusions were:
  • GM maize grain yield was increased by up to 24.5% (due to reduced pest damage) 
  • broad compositional analyses of the grain wasn't altered by genetic transformation 
  • fungal toxin contamination was around one-third lower in GM maize (due to reduced ear damage by pests ) 
  • GM maize had no substantial effect on abundance or diversity of the non-target insect community 
  • inexplicably GM maize had an increased decomposition rate 
Does this lay your paranoid GM worries categorically to rest?

Have another look at those five worries: your health, the lack of safety testing on humans, increased human exposure to pesticides, reduced crop diversity, and effects on wildlife.

Apart from fungal toxins, which are an ongoing, well-known serious health risk in maize only partially linked to pest damage in the field (see Note), and consequently subject to regulation, your worries (No.1 and 2) about health effects and lack of appropriate safety testing have not been addressed.

Where crops generate their own pesticides, or are designed to tolerate spraying with pesticides, human exposure to these will, inevitably, be increased (worry No.3). However, no pesticide measurements were reported. 

Crops diversity crash caused by increasing biotech industry control of the seed supply (worry No.4) has certainly not been considered.

In the short-term, animal wildlife was minimally altered in the GM maize stands investigated. This would be good news for worry No.5 if it wasn't for the 'short-term' bit and for the fact that the studies weren't comprehensive.

To sum up, the GM crop data fed into the meta-analysis weren't actually collected with the aim of calming consumer concerns about safety. It's likely they were more to do with the process of commercialisation of the GM maize (more on this below).

Since the paper passed peer-review, the meta-analysis must be acceptable, and the researchers noted numerous limitations to their data. However, the title "Impact of genetically engineered maize on agronomic, environmental and toxicological traits: a meta-analysis of 21 years of field data" is misleading. It sounds like a study of the impacts GM maize has actually achieved in farming, on the environment and on food toxicity in the course of the 21 years it's been in use.

In fact, the 76 sets of data fed into the meta-analysis came from field trials of a type carried out early in the stages of commercial development of a GM crop: they're neither large-scale nor long-term; they're not designed to reveal emerging pest resistance, nor novel pests moving into the Bt-created void, nor sub-chronic harm to wildlife, nor adverse nutritional qualities all of which can become a reality over time.

As Testbiotech points out "According to data from industry, 100 (genetically engineered) maize events are grown worldwide. Most of them produce Bt toxins. Currently, around a dozen Bt toxins are produced in genetically engineered plants, all belonging to different classifications". These GM events may include herbicide-tolerance traits and stacks of two to six artificial genes. The 76 studies certainly couldn't have included all of the GM events on the market, and since successful traits from the laboratory are usually bred into elite strains for commercialisation, it's possible that not one of the 76 tested a form of GM maize actually in your food chain or environment. 

To the cynical, most (possibly all) these field trials were sponsored by industry to ease the GM trait's way through the approval process and to aid marketing. If the results weren't positive, would they have been published?


It's intriguing that the meta-analysis showed an increased decomposition rate in GM maize plant-matter compared with its closest non-GM relatives. The best explanation the researchers could offer was that there might be differences in the proportion and composition of plant stalks and leaves in the decomposing residue. This suggests that the GM maize plants and their closest non-GM relatives have fundamental differences in their physiology and growth. Compositional factors altering the GM plant's susceptibility to the microbes instrumental in decomposition could also play an interesting part.

At least two sensational reports on the GM maize meta-analysis in the popular press misunderstood the nature of the data: for example, one described the paper as an analysis of approximately 6,000 studies; one claimed the study proved that the risk of GM maize was small, the other claimed it proved GM maize is not only safe but has health benefits. Neither understood the strengths and limitations of the scientific field trials used, nor that 'impact' of GM maize wasn't actually what was investigated.

The moral is beware of dramatic pro-GM 'scientific' pronouncements by writers whose expertise lies in unrelated areas such as, law economics, consumer markets and journalism (especially if there's a claimed unique ability to speed read through dense scientific journals).

Such poorly researched articles only serve to heighten public distrust of GM.


Note on mycotoxins

Mycotoxin contamination of maize is generally climate-dependent and is associated with poor practice during harvesting, storage, transport and processing and with insufficient pest control at all stages. Poor sanitary conditions during handling exacerbate mycotoxin presence. In the field, burial of the crop residue and good crop rotation can be used to reduce mycotoxin contamination.


  • Elisa Pellegrino, et al., 15th February 2018, Impact of genetically engineered maize on agronomic, environmental and toxicological traits: a meta-analysis of 21 years of field data, Nature Scientific Reports
  • Meta-analysis of data from GE maize fails reality check, Testbiotech, 20.03.18
  • Media Reporting on New GMO Safety Study is Deceptive: Flawed Meta-Analysis Debunked,, 22.03.18
  • Nonprofit says new GM corn report is sleight of hand, GM Watch 3.04.18
  • Omri Ben-Shahar, The Environmentalist Case In Favor Of GMO Food,, 26.02.18
  • Dana Dovey, GMO corn is safe and even has health benefits, analysis of 6,000 studies concludes,, 22.02.18
  • Dana Dovey profile in
  • Zengran Liu, et al., 2016, Factors controlling mycotoxin contamination in maize and food in the Hebei province, China, Agronomic Sustainable Development

No comments:

Post a comment

Thanks for your comment. All comments are moderated before they are published.