Glyphosate accounted for 62% of herbicide use in Brazil between 2009 and 2016, and it was especially used in soybean production. In 2017, it represented 30% of global herbicide use. It is estimated that since the 1990s, following the creation of seeds resistant to this substance, the use of this product has increased around 15-fold. Furthermore, according to Data Intelligence, the use of glyphosate on crops is expected to grow 800% by 2025. In this context, a study by Fundação Getulio Vargas found that this herbicide has effects on human health and identified an increase in infant mortality in locations close to where glyphosate has been applied.

The study indicates that between 2000 and 2010, there was a 5% increase in the infant mortality rate, which corresponds to around 503 deaths per year, among infants whose mothers lived close to where glyphosate was applied, more specifically, in the path of water flow carrying the substance applied to crops into rivers, streams and wells. The period analyzed in the research coincides precisely with the time when there was a sharp rise in the use of glyphosate.

Despite these indications found by the scientists, Rudi Rocha, a researcher at Fundação Getulio Vargas’ Sao Paulo School of Business Administration (FGV EAESP) and co-author of the study, says that this figure probably underestimates the overall effect of glyphosate use on human health.

“There is concern about the subclinical toxicity of these substances on people in general, who are not subject to direct exposure, but exposed to low concentrations through ingestion of contaminated water or food,” the researcher explains. He also points out that this type of intoxication is just the tip of the iceberg, as pesticides like this are capable of causing a wide range of asymptomatic effects at very low levels of exposure, which do not produce obvious signs and symptoms.

“The population affected by this subclinical toxicity could be much larger than the group affected by direct exposure,” Rocha warns. He also points out that it is not clear how this indirect contamination could have significant effects on human health, which is why he highlights the importance of studies like this to understand the consequences of pesticide use in general, in order to contribute to public debate on the subject, which until now has been dominated by economic interests and scientific controversies.

According to the researcher, producers of this herbicide often claim that the physical and chemical conditions of the product are good, as it has a short half-life, which causes glyphosate to break down into another, less harmful substance. However, there is now scientific evidence indicating the presence of the product in water samples from rivers and streams, among other places, some distance from the location where the chemical was applied.

Conducting the experiments

Rocha says that according to laboratory tests, human embryos are particularly sensitive to environmental conditions and glyphosate. This product, when applied to crops to eliminate weeds, could potentially affect placental cells and fetuses. To test the level of this effect, Rocha made a series of econometric estimates, together with researchers Mateus Dias from the Catholic School of Business and Economics in Lisbon and Rodrigo Soares from Insper, a higher education institute in Sao Paulo. The group analyzed municipal databases covering all the municipalities in Brazil’s South and Midwest regions.

In order to capture the effect of glyphosate on human health, the group first had to eliminate correlated factors. “It’s very difficult to measure the impact of pesticides on human health for many reasons. It’s complicated, because many regions that use these products start to grow economically and connect with other regions, generating immigration, and so on. So, it’s a challenge to specifically isolate the effect of glyphosate on health,” says Rocha.

To gauge this product’s effect on human health without socioeconomic interference, the researchers used an econometric method based on a statistical technique known as instrumental variable analysis, which can measure natural suitability for glyphosate at a local level, without being correlated with other characteristics.

“We explored the idea that if it is used in a specific area, this herbicide may contaminate the soil and, through water flow, be carried downstream, affecting people who consume water far from where the pesticide was used,” the researcher explains.

The study used databases of Brazilian river basins to investigate the flow of water in each of them and link this information to birth data. In addition, precipitation data was used to identify the level of rainfall and assess whether the water flow was capable of carrying glyphosate to other regions.

“The idea was as follows: in a given location where the use of the substance increased, which occurred mainly after the use of genetically modified soybean seeds resistant to glyphosate was permitted, we analyzed the impact on children’s health in populations living in nearby places, considering the underground and surface hydrographic flow. In other words, if a farmer applied glyphosate to his crops, we looked at the downstream municipalities, where that potentially contaminated water would pass through,” Rocha says.

When water flow is fatal

When investigating the presence of glyphosate in this water flow, the researchers found a deterioration in infants’ health precisely in areas close to and downstream from where there was an increase in the use of glyphosate. This phenomenon has increased considerably since 2004, when the law was changed to allow the use of genetically modified soybean seeds.

In addition, the greater the volume of rain in the area where the product is applied, and the steeper the terrain, the greater the spread of the substance to the surrounding areas and, consequently, the greater the impact on child mortality. “When it rains more, and where the terrain is steeper and has a higher level of erodibility, the rain takes the product to other locations more quickly,” says Rocha.

As well as infant mortality, the researchers also identified an increase in premature births and low birth weight in locations downstream from areas where glyphosate is used on crops.

To ensure that the harm to human health came from this specific herbicide and not from other substances present in the water, the researchers carried out a series of robustness tests to assess the accuracy and stability of the estimates. “We used data from water treatment plants to check whether substances other than glyphosate were present, and no traces of other products were found,” says Rocha.

Finally, the researcher reiterates that the mortality profile is consistent with what would be expected from exposure to glyphosate during pregnancy: 56% of the total effect came from perinatal conditions and 19% from respiratory conditions. He adds that scientists have recently re-examined claims that glyphosate is a safe pesticide with little or no effect on human health.

“There was already some more direct evidence of glyphosate’s effect on human health, given that on many occasions, farmers have used this product and fallen ill after handling it. However, there was no information on how this substance could affect the health of the surrounding population. As these results were unknown when the current regulations were established, a new discussion needs to start on the regulatory framework for the use and handling of glyphosate-based herbicides,” he concludes

To read the full study, click here.

For more information about FGV EAESP, click here.