If you look at Figure 2. you will see the relative risks are mostly less than 1.0. In other words, more radiation means less cancer. Dr. Nair himself fitted a straight line to his data using ordinary least squares and came with a slope of -0.13/Gy.
BTW, the text d0es not match the figure exactly. It should says something like the blue spike shows the Committee is nearly certain that the slope is between -0.12 and -0.15. I was playing around with priors trying to find a density that best matched the Beir VII Committees prior numbers.
I'm not sure what happened here. It appears I answered your earlier question which you replaced before I could respond. That's not good etiquette.
To answer this question, I dont think so as I said in the piece. To make this case, you would carefully have to correct for all the confounding factors. Nair et al tried to do this for a few possibilities, (bibi smoking, religion, etc) but what's really different about the various bins is how close to the shore they are. I'd make a wild guess that the people close to shore are a little wealthier than the people inland. If so that would completely dominate any effect on the tiny differences in dose rate.
In my view, hormetians are like LNTers in that they make low dose rate radiation a far more powerful agent than it is.
Sorry about changing my reply. Shortly after posting it, I realized I hadn’t read your post carefully enough.
So many other agents exhibit a hormesis response, why not radiation? Do you know of any agent that doesn’t exhibit hormesis? I get the impression that you think hormesis is a bridge too far; that politically it’s just too much to ask. You’re probably right. Plus there is probably not enough data to define any response curve with any accuracy. But I thought there some evidence that higher background radiation was beneficial.
Hormesis is real. But the clearest examples are when a small dose rate pulse is followed by a much larger pulse. The first dose stimulates repair processes which end up reducing the harm associated with the big dose rate pulse. But continually over-stimulating repair processed has to reach saturation pretty quickly. Otherwise a release is good for you and we should pay NPP to spread some radiation around routinely. If they won't do it, put a radon source in your basement.
The main epidemiological argument for hormesis from chronic dose rate profiles in Cohen's study. But Cohen's data is subject to the ecological fallacy of claiming that you can sum doses to get the sum of the harm. This is true only if the dose response curve is linear. So while Cohen's study is a compelling falsification fof LNT which claims you can do exactly that, it is far from compelling for a non-LNTer. Cohen himself made that point.
I did not answer your other question. Of course, there are agents that exhibit no hormetic effect. Lead is one. Just about any really heavy metal including uranium's chemical toxicity.
But yeah. Too little radiation is bad for you. But I think the overall evidence is that too little is well below normal background.
The important point here is not whether hormesis exists, but getting rid of LNT. The replace LNT we much ahve a fully defiwend replaccement. That means a computer program that can convert any dose rate profile into a prediction of cancer incidence. Until the hormetians show me that program, I pay them very little attention.
The real problem is when hormetians argue against SNT on the grounds that it is not "good enough" but then don't come up with that replacement program. That plays into the hands of the LNTers. Such hormetians have blood on their hands, although I don't think they are aware of it.
I wouldn't push the Ramsar experience very hard. I don't think we have anything like Nair/Amma study of the Karunagappally cohort in terms of numbers and effort and care for Ramsar. And on this site we don't talk about annual doses. Annual has no biological meaning. We talk daily doses because a day is much closer to the cell DNA repair period. The Ramar numbers you quote are 0.03 mSv/day to 0.4 mSv/d. The quite large high end cohort in Nair's study averaged about 0.2 mSv/d which means some were considerably higher. The Karunagappally data is far more compelling than the Ramsar.
One of my long running frustrations is the binning of data into arbitrary cohorts. That is confuscation at its worst, introducing all manor of bias and requiring further arbitrary artifacts such as error bars to be deployed. Sadly everyone and their grandmother seems to be doing it all the time.
If the data is good, then use a point cloud. My monitor contains over 2 million pixels. Plenty enough for even the most ambitious studies.
Since all of this is obvious, we can deduce that either the data is no good, or the scientists are no good. Maybe both!
I agree. Make your raw data available. I thought that was a basic rule of scientific integrity.
Nair has not made his raw data available, only the number of people in each bin, N_bin,
and the 95% confidence interval for that bin, Therefore, for the data, I generated N_bin random samples for each bin assuming a Normal with a mean of the bin excess relative risk and standard deviation of 1/4 Nair's 95% confidence interval. This synthesized data is far from ideal --- everybody in each bin gets the average dose for that bin --- but in this case I'm confident the overall results would be the same as if I had the real data.
This failure to disclose data is endemic to the radiation protection racket. I know of only one exception: Cohen. I explicitly asked the NIH for their Chernobyl thyroid data, and they told me I did not qualify to get "sensitive" data.
“… the kind of dose rate profiles (DRP) incurred by the Karunagappallians decrease your chance of getting cancer.”
Could this be hormesis in action?
Tom,
If you look at Figure 2. you will see the relative risks are mostly less than 1.0. In other words, more radiation means less cancer. Dr. Nair himself fitted a straight line to his data using ordinary least squares and came with a slope of -0.13/Gy.
BTW, the text d0es not match the figure exactly. It should says something like the blue spike shows the Committee is nearly certain that the slope is between -0.12 and -0.15. I was playing around with priors trying to find a density that best matched the Beir VII Committees prior numbers.
I'm not sure what happened here. It appears I answered your earlier question which you replaced before I could respond. That's not good etiquette.
To answer this question, I dont think so as I said in the piece. To make this case, you would carefully have to correct for all the confounding factors. Nair et al tried to do this for a few possibilities, (bibi smoking, religion, etc) but what's really different about the various bins is how close to the shore they are. I'd make a wild guess that the people close to shore are a little wealthier than the people inland. If so that would completely dominate any effect on the tiny differences in dose rate.
In my view, hormetians are like LNTers in that they make low dose rate radiation a far more powerful agent than it is.
Sorry about changing my reply. Shortly after posting it, I realized I hadn’t read your post carefully enough.
So many other agents exhibit a hormesis response, why not radiation? Do you know of any agent that doesn’t exhibit hormesis? I get the impression that you think hormesis is a bridge too far; that politically it’s just too much to ask. You’re probably right. Plus there is probably not enough data to define any response curve with any accuracy. But I thought there some evidence that higher background radiation was beneficial.
Tom,
Hormesis is real. But the clearest examples are when a small dose rate pulse is followed by a much larger pulse. The first dose stimulates repair processes which end up reducing the harm associated with the big dose rate pulse. But continually over-stimulating repair processed has to reach saturation pretty quickly. Otherwise a release is good for you and we should pay NPP to spread some radiation around routinely. If they won't do it, put a radon source in your basement.
The main epidemiological argument for hormesis from chronic dose rate profiles in Cohen's study. But Cohen's data is subject to the ecological fallacy of claiming that you can sum doses to get the sum of the harm. This is true only if the dose response curve is linear. So while Cohen's study is a compelling falsification fof LNT which claims you can do exactly that, it is far from compelling for a non-LNTer. Cohen himself made that point.
I did not answer your other question. Of course, there are agents that exhibit no hormetic effect. Lead is one. Just about any really heavy metal including uranium's chemical toxicity.
But yeah. Too little radiation is bad for you. But I think the overall evidence is that too little is well below normal background.
The important point here is not whether hormesis exists, but getting rid of LNT. The replace LNT we much ahve a fully defiwend replaccement. That means a computer program that can convert any dose rate profile into a prediction of cancer incidence. Until the hormetians show me that program, I pay them very little attention.
The real problem is when hormetians argue against SNT on the grounds that it is not "good enough" but then don't come up with that replacement program. That plays into the hands of the LNTers. Such hormetians have blood on their hands, although I don't think they are aware of it.
I wonder what BEIR thinks about the people in Ramsar, Iran. Some get annual effective doses ranging from 10 to 130 mSv.
Phil.
I wouldn't push the Ramsar experience very hard. I don't think we have anything like Nair/Amma study of the Karunagappally cohort in terms of numbers and effort and care for Ramsar. And on this site we don't talk about annual doses. Annual has no biological meaning. We talk daily doses because a day is much closer to the cell DNA repair period. The Ramar numbers you quote are 0.03 mSv/day to 0.4 mSv/d. The quite large high end cohort in Nair's study averaged about 0.2 mSv/d which means some were considerably higher. The Karunagappally data is far more compelling than the Ramsar.
"You must become a hormetian LNTer" is borderline epitaph worthy
Here's a nice tutorial on Bayesian statistics. I'm working through it. https://statswithr.github.io/book/the-basics-of-bayesian-statistics.html
One of my long running frustrations is the binning of data into arbitrary cohorts. That is confuscation at its worst, introducing all manor of bias and requiring further arbitrary artifacts such as error bars to be deployed. Sadly everyone and their grandmother seems to be doing it all the time.
If the data is good, then use a point cloud. My monitor contains over 2 million pixels. Plenty enough for even the most ambitious studies.
Since all of this is obvious, we can deduce that either the data is no good, or the scientists are no good. Maybe both!
I agree. Make your raw data available. I thought that was a basic rule of scientific integrity.
Nair has not made his raw data available, only the number of people in each bin, N_bin,
and the 95% confidence interval for that bin, Therefore, for the data, I generated N_bin random samples for each bin assuming a Normal with a mean of the bin excess relative risk and standard deviation of 1/4 Nair's 95% confidence interval. This synthesized data is far from ideal --- everybody in each bin gets the average dose for that bin --- but in this case I'm confident the overall results would be the same as if I had the real data.
This failure to disclose data is endemic to the radiation protection racket. I know of only one exception: Cohen. I explicitly asked the NIH for their Chernobyl thyroid data, and they told me I did not qualify to get "sensitive" data.