Welp, here it is -- the study that slams the book shut on the whole thing. Hardly worth discussing now, as it's settled science. Back to work, everyone.
Care to elaborate? I'm following that from a distance and my impression was, that until now? Glyphosate was considered safe i.e. German BfR (risk assessment institute) said it there no big risks in the last EU talks... a lot of sceptics are also saying this.
So - basically this proves that these were wrong (or misled) and exposure to Glyphosat causes cancer?
Note the distinction between "glyphosate" and "glyphosate-based herbicide". The former was well studied and presumed safe. My understanding is it is the actual formulations that are being claimed as unsafe now. I.e., if you ship your safe herbicide in an unsafe slurry you get cancer.
Interesting. In the end in reality everyone is using some kind of "glyphosate-based herbicide" through and the whole discussion is about an chemical in isolation. Talk about misleading the public discourse :/
I've read that in practice the ingredients of "glyphosate-based herbicide" are often treated as trade-secrets and are not public, does anyone here have more information?
> Please don't impute astroturfing or shillage. That degrades discussion and is usually mistaken. If you're worried about it, email us and we'll look at the data.
" Three of the study authors were tapped by the EPA as board members for a 2016 scientific advisory panel on glyphosate. The new paper was published by the journal Mutation Research /Reviews in Mutation Research, whose editor in chief is EPA scientist David DeMarini.
The study’s authors say their meta-analysis is distinctive from previous assessments. “This paper makes a stronger case than previous meta-analyses that there is evidence of an increased risk of NHL due to glyphosate exposure,” said co-author Lianne Sheppard, a professor in the Environmental and Occupational Health Sciences department at the University of Washington. “From a population health point of view there are some real concerns.”
Sheppard was one of the scientific advisers to the EPA on glyphosate and was among a group of those advisers who told the EPA that it failed to follow proper scientific protocols in determining that glyphosate was not likely to cause cancer. “It was wrong,” Sheppard said of the EPA glyphosate assessment. “It was pretty obvious they didn’t follow their own rules. “Is there evidence that it is carcinogenic? The answer is yes.”
An EPA spokesperson said: “We are reviewing the study.”
It should be noted that this is about 'most highly exposed' farmers who were (presumably) exposed to large amounts of Roundup, unprotected. Regular consumers don't get these exposures in their lifetimes.
This has been shown before, which is why OP's paper it a meta analysis of available papers..
Likewise it focuses only on a population of people who are exposed to a lot of pesticides, herbicides, fungicides, etc. and doesn't control for just glyphosate-based herbicide exposure. The incidence of non-Hodgkin lymphoma could be correlative with this population for a number of causes, so further investigation would be good.
This is a review, not original research. It seems to also conduct a meta-analysis as part of this, however that is somewhat unusual for a review. In general, both the review and meta-analysis quality is dependent on the quality of the source material. Garbage in and garbage out.
"Some epidemiological studies have reported an increased risk of NHL in GBH-exposed individuals [[15], [16], [17]]; however, other studies have not confirmed this association [18,19]." This is why we should approach this with skepticism.
"the 2018 AHS update [24] contributes 11-12 additional years of follow-up with over five times as many NHL cases" If we look at the AHS updates conclusions they say "In this large, prospective cohort study, no association was apparent between glyphosate and any solid tumors or lymphoid malignancies overall, including NHL and its subtypes." so why are we adding in older controversial studies with less power and sample size?
The other studies used in the meta-analysis are confounded by the use of up to 40 different pesticides. The confidence intervals on those case-control studies range from 1.0-6.0 odds ratio for glyphosate significance with its association with NHL, so basically all across the board. For herbicides with known worse toxicity, the effect and confidence intervals is much more robust with odds ratios bottoming out around 1.5. Again, this relates to garbage in/garbage out, if we are combining a bunch of confounded studies with one glyphosate specific one (the AHS one quoted above), it might bias the results.
Tables 1-4 - background on the N=6 studies in the meta-analysis
Figure 2 tells the story, huge confidence intervals, and the older studies are the ones which show the effect. The larger glyphosate specific studies show no effect.
Conclusion :
I don't find this to be compelling evidence that glyphosate exposure is linked to NHL. Recent, specific studies find no effect and adding in older, less specific research will only bias and confound that result.
>Because if meta-studies only get to include the single most recent and largest study, they wouldn’t be very ‚meta‘, would they?
I'm saying the earlier are clearly non-specific to glyphosate. The 2015 and 2018 AHS are targeted to glyphosate specifically and with larger samples of cases, and thus warrants more consideration. If your meta-study can only be done by including specious sources, then maybe that should be questioned.
>I love how you’re trying to appear all scientificatious while really just saying „some say A, others say B. No way to know. Be sceptical!“
Yes that is exactly what I am saying.
The results of multiple independent researchers are inconclusive, thus anyone claiming clear evidence of anything should be have to prove their point very robustly, particularly when the issue is charged with social/political bias.
I didn't follow the whole glyphosate-debate for a while. When I did, some years ago, it seemed as if there were mostly studies linking it to cancer etc in vitro or in rodents etc, not so much or even none in humans. Then last year when a judge in the US ruled against Monsanto some people were like 'hey here's your evidence that glyphosate causes cancer' which is obviously wrong, hnce the many replies saying there were no conclusive studies in humans linking it to cancer. So, did this change now?
>it seemed as if there were mostly studies linking it to cancer etc in vitro or in rodents etc, not so much or even none in humans.
context:
(14 January 2019 )
As one reviewer put it: “The article transparently lays out not just that the EPA and IARC came to different conclusions about the genotoxicity of glyphosate-based herbicides, but how this result occurred and its impact on the overall conclusions about its carcinogenicity. The analyses contained in this article and accompanying text enhances the understanding of the state of the science of the potential carcinogenicity of glyphosate-based herbicides and gaps in understanding that future studies may help to resolve. As such, it is an important contribution to the literature”
-----------
more ( & Open )
-----------
Editorial:
"Some food for thought: a short comment on Charles Benbrook´s paper “How did the US EPA and IARC reach diametrically opposed conclusions on the genotoxicity of glyphosate-based herbicides?” and its implications"
Whether or not this will hold up over time (I'm skeptical, but I'll keep an open mind), it's appalling that a report with potentially enormous implications for agriculture and public health is locked up behind the Elsevier paywall where I can't read it. I can't believe it's 2019 and I'm still complaining about this.
From just reading the abstract, relative risk of 1.41 for the highest exposure groups when available in each study isn't something I would consider "risk". Even if it is statistically significant, the effect size is pretty small.
Cigarette smoking is a long-known risk and thus has warning labels on every pack. Of course the RR is higher. Comparing that known large risk factor to make another seem smaller by comparison is at best disingenuous.
30 comments
[ 2.6 ms ] story [ 77.2 ms ] threadSo - basically this proves that these were wrong (or misled) and exposure to Glyphosat causes cancer?
I've read that in practice the ingredients of "glyphosate-based herbicide" are often treated as trade-secrets and are not public, does anyone here have more information?
A single study does not a scientific consensus make.
Well-accepted scientific facts are formed from dozens of studies, often with conflicting results (both positively and negatively).
I don't think you understand how science works.
https://news.ycombinator.com/newsguidelines.html
"""
" Three of the study authors were tapped by the EPA as board members for a 2016 scientific advisory panel on glyphosate. The new paper was published by the journal Mutation Research /Reviews in Mutation Research, whose editor in chief is EPA scientist David DeMarini.
The study’s authors say their meta-analysis is distinctive from previous assessments. “This paper makes a stronger case than previous meta-analyses that there is evidence of an increased risk of NHL due to glyphosate exposure,” said co-author Lianne Sheppard, a professor in the Environmental and Occupational Health Sciences department at the University of Washington. “From a population health point of view there are some real concerns.”
Sheppard was one of the scientific advisers to the EPA on glyphosate and was among a group of those advisers who told the EPA that it failed to follow proper scientific protocols in determining that glyphosate was not likely to cause cancer. “It was wrong,” Sheppard said of the EPA glyphosate assessment. “It was pretty obvious they didn’t follow their own rules. “Is there evidence that it is carcinogenic? The answer is yes.”
An EPA spokesperson said: “We are reviewing the study.”
""" https://www.theguardian.com/business/2019/feb/14/weed-killin...
You can read the whole paper on sci hub
This is a review, not original research. It seems to also conduct a meta-analysis as part of this, however that is somewhat unusual for a review. In general, both the review and meta-analysis quality is dependent on the quality of the source material. Garbage in and garbage out.
"Some epidemiological studies have reported an increased risk of NHL in GBH-exposed individuals [[15], [16], [17]]; however, other studies have not confirmed this association [18,19]." This is why we should approach this with skepticism.
"the 2018 AHS update [24] contributes 11-12 additional years of follow-up with over five times as many NHL cases" If we look at the AHS updates conclusions they say "In this large, prospective cohort study, no association was apparent between glyphosate and any solid tumors or lymphoid malignancies overall, including NHL and its subtypes." so why are we adding in older controversial studies with less power and sample size?
The other studies used in the meta-analysis are confounded by the use of up to 40 different pesticides. The confidence intervals on those case-control studies range from 1.0-6.0 odds ratio for glyphosate significance with its association with NHL, so basically all across the board. For herbicides with known worse toxicity, the effect and confidence intervals is much more robust with odds ratios bottoming out around 1.5. Again, this relates to garbage in/garbage out, if we are combining a bunch of confounded studies with one glyphosate specific one (the AHS one quoted above), it might bias the results.
Tables 1-4 - background on the N=6 studies in the meta-analysis
Figure 2 tells the story, huge confidence intervals, and the older studies are the ones which show the effect. The larger glyphosate specific studies show no effect.
Conclusion :
I don't find this to be compelling evidence that glyphosate exposure is linked to NHL. Recent, specific studies find no effect and adding in older, less specific research will only bias and confound that result.
Because if meta-studies only get to include the single most recent and largest study, they wouldn’t be very ‚meta‘, would they?
> however, other studies have not confirmed this association [18,19]." This is why we should approach this with skepticism.
I love how you’re trying to appear all scientificatious while really just saying „some say A, others say B. No way to know. Be sceptical!“
I'm saying the earlier are clearly non-specific to glyphosate. The 2015 and 2018 AHS are targeted to glyphosate specifically and with larger samples of cases, and thus warrants more consideration. If your meta-study can only be done by including specious sources, then maybe that should be questioned.
>I love how you’re trying to appear all scientificatious while really just saying „some say A, others say B. No way to know. Be sceptical!“
Yes that is exactly what I am saying.
The results of multiple independent researchers are inconclusive, thus anyone claiming clear evidence of anything should be have to prove their point very robustly, particularly when the issue is charged with social/political bias.
context:
(14 January 2019 )
As one reviewer put it: “The article transparently lays out not just that the EPA and IARC came to different conclusions about the genotoxicity of glyphosate-based herbicides, but how this result occurred and its impact on the overall conclusions about its carcinogenicity. The analyses contained in this article and accompanying text enhances the understanding of the state of the science of the potential carcinogenicity of glyphosate-based herbicides and gaps in understanding that future studies may help to resolve. As such, it is an important contribution to the literature”
-----------
more ( & Open )
-----------
Editorial:
"Some food for thought: a short comment on Charles Benbrook´s paper “How did the US EPA and IARC reach diametrically opposed conclusions on the genotoxicity of glyphosate-based herbicides?” and its implications"
https://enveurope.springeropen.com/articles/10.1186/s12302-0...
-----------
more ... more ... ( & Open )
-----------
(14 January 2019)
"How did the US EPA and IARC reach diametrically opposed conclusions on the genotoxicity of glyphosate-based herbicides?"
https://enveurope.springeropen.com/articles/10.1186/s12302-0...
For comparison, the RR for cigarette smoking and (small cell) lung cancer is estimated to be around 21 (REF: http://cebp.aacrjournals.org/content/14/9/2125)