Anti-fluoridationists exploit infant deaths by fiddling statistics

A useful reminder of how activists lie with statistics

The local anti-fluoride people have not stopped rabbiting away with their “science” – searching for anything bad they can argue is caused by community water fluoridation. The latest claim they make – fluoridation is responsible for infant deaths!

And they have some statistics to “prove” this. The table below presented by well-known US anti-fluoride activist, Karen Favazza Spencer, in her article America First – Chemical Warfare on Infants:

That table is simply lifted from a conspiracy style web page Why Pregnant Women Should Not Drink Fluoridated Water.

OK – these statistics might fool some people – especially if you have a bias to confirm. But the more critical person might pick up that these figures are most likely cherry-picked and want to see the full data set or some analysis of the data.

It’s not hard to find this data as there are tables of all sorts of things for US people organised by state. So, is there a relationship between infant deaths in each state and the extent of fluoridation in each state?

There actually does appears to be one at first sight – here is the graph of the data for infant deaths in 213 plotted against the extent of fluoridation in 2012.

ID-fl

But, just a minute – it is not actually statistically significant (p=0.106) and would account for only about 5% of the variance in infant deaths. Fluoridation is certainly not the main factor – and probably involved at all if other factors are considered.

Here I will just take into account the influence of state elevation – because I know from previous work that fluoridation extent is related to mean state elevation (see ADHD linked to elevation, not fluoridation).

Here is that relationship for the extent of fluoridation in 2012:

Fl-elev-2012

So, the extent of fluoridation in each state is related to mean state elevation and this relationship is statistically significant (p=0.005). Actually not surprising as the larger and older cities where fluoridation might be expected are generally situated at lower elevations for a number of reasons.

But what is the relationship between infant death and mean state elevation? Well, it is stronger than for the extent of fluoridation (p=0.002). Elevation accounts for about 18% of the variance in infant deaths in 2013.

ID-ele

Finally, let’s combine both elevation and extent of fluoridation into a multiple regression and see what the relationship when both factors are combined.

This multiple regression shows a statistically significant association (p=0.007) of the extent of infant deaths in each state in 2013 with the mean state elevation. However, there is no statistically significant association (p=0.592) with the extent of state fluoridation.

So while infant deaths could be explained by mean state elevation and most probably one or more other factors, they certainly are not explained by the extent of fluoridation. Not at all!

Preterm birth and conspiracy theory

In her article, Karen Favazza Spencer makes the bald claim “Fluoridation is positively correlated with preterm birth and increased death rates by state “ – again citing from the conspiracy style web page mentioned above.

Sure, that page makes that claim – “Domestic water fluoridation was independently associated with an increased risk of PTB [preterm birth].”  But that is hardly credible evidence because that page goes on the say:

“This study was never published nor was any follow-up research done, despite the fact that 2 years earlier, the US Institute of Medicine reported: ‘Those born preterm have an appreciable risk of long-term neurological impairment and developmental delay.'”

Strange! It is not hard to find data for preterm birth. In fact, here it is for 2014 compared with the extent of state fluoridated in 2012:

Clearly, there is no association between preterm births and extent of fluoridation. Yet that web page claimed there was and that the information had been suppressed!

I guess that is another way ideologically motivated activists “prove” these sort of things – invoke a conspiracy theory to claim a relationship exists but the data is suppressed.

So, once again the lesson is – never take at face value the claims made by anti-fluoridation activists – no matter how “sciency” their information looks or what data they invoke to “prove” them.

Always check such claims for yourself.

Similar articles

Advertisements

110 responses to “Anti-fluoridationists exploit infant deaths by fiddling statistics

  1. Nothing new there. Ms Spencer is a master at quoting Papers that have nothing to do with the argument she is trying to defend.
    So the point, cited to check all anti fluoride links is so relevent

    Like

  2. More water can be consumed in pregnancy but less probably at higher altitudes which are cooler.

    Rather than extent of fluoridation need to work with extent of fluoridated water consumed as that increases fluoride intake.

    Elevated towns may be smaller and use sodium fluoride for fluoridation. Do they add lime in similar quantities or maybe alkalinise with sodium carbonate or hydroxide?
    And we already came across a sodium fluoride fluoridation health problem didn’t we?

    And its all crap unless you allow for magnesium

    Like

  3. Brian, I am afraid your comment is all crap – it does not even refer to the point of my article – the misuse of statistics.

    A proper look at these shows absolutely no association of infant death or preterm births with fluoridation. None at all!

    Hence your attempted diversion.

    Like

  4. I was saying the things a “proper look” should consider.

    Like

  5. So your comment is a criticism of FFNZ and of Karen Favazza Spencer, Brian?

    Well, why didn’t you say. I think my criticism was fundamental.

    Like

  6. “And it’s all crap.”

    That sentence would have been accurate if it stopped there.

    Like

  7. Writers should be encouraged to take many possible variables into account.

    Altitude affects air pressure which is not good for pre-term births. It also affects UVB which may be good for pregnancy via vitamin D.

    If some individuals be more susceptible to fluoride via COMT variant I assert it be not fair to write them off using the excuse altitude shows a greater affect over the whole population.

    Like

  8. Even low air pressure at altitude does not affect all individuals the same.

    Here compare Tibet: the “original” inhabitants vs the newly arrived Han Chinese as regards neonatal troubles:

    http://www.nejm.org/doi/full/10.1056/NEJM199511093331903#t=article

    The Dunedin Study which TV series Ken asked us to watch related of the COMT variant and how it did not have to be followed by antisocial characteristics but might even be banked on for special ability. It would be interesting to look at advanced rugby players and their COMT variants. They may be more affected by fluoride.

    I assert it be not the simple affect on all the sports people that Stuartg is suggesting.

    Like

  9. Ken I think your chart of infant deaths vs average state elevation should be segmented with an increase in the rate to about 7,000 feet as air pressure decreases followed by a decrease as the increased UVB effect shows.

    There are a few high outliers possibly.

    Need also to investigate residence in high rise buildings and note the Trumps live on the 66th floor where the air pressure must be a percent lower than ground level.

    (I can find my thinking to be not so clear even on a plane ascending to 18,000 feet where pressurisation keeps it like 900 – about 3,000 feet.)

    Like

  10. soundhill,

    “the simple affect on all the sports people that Stuartg is suggesting”

    If you read back, you will see that I’ve made no mention at all about the mental health of sports people. Oh, wait, it’s yet another spelling mistake that wasn’t picked up by proof reading.

    I suggest that you read back over the last few years and see exactly who has suggested that CWF has an effect, either simple or complex, on sports people.

    All I’ve ever done is ask that person to provide evidence why they believed the coincidences they’d found weren’t. They haven’t provided any.

    Like

  11. 14th June last year TV1 played a Dunedin Study episode which dealt the forms of the MAOA gene. 30% of people have the weak form and it is claimed they can become antisocial if maltreated up to age 11. Professor James Fallon who found 7 convicted murderers in his father’s family tree got his own brain scanned and it showed the problem pattern. But his mother had treated him well and his energy chanelled him to be a professor if you call that mental health.

    Stuartg: “If you read back, you will see that I’ve made no mention at all about the mental health of sports people.”

    I imagine top rugby players need controlled aggression if you call that mental health.

    I have been talking about top players but Stuartg wants me to extend the study to all players, suggesting that any problem of fluoridation on top players should affect all players.

    Like

  12. soundhill,

    “I have been talking about top players”

    I suggest that you read back.

    Let us know who extended their fantasy to lower grade players by first mentioning lower grade rugby teams, such as Timaru and Petone.

    Let us know who first mentioned football teams in the lower UK leagues such as Birmingham City.

    In fact, you could let us know who first raised the fantasy that CWF could reduce sports players performance by “more than 10%”.

    Then let us know who hasn’t been able to demonstrate significance in the tiny number of coincidences he’s been able to cherry pick among the thousands (millions?) of results available for him to peruse.

    Like

  13. soundhill,

    FYI, affect ≠ effect.

    Like

  14. soundhill,

    “Stuartg wants me to extend the study to all players, suggesting that any problem of fluoridation on top players should affect all players.”

    Read back.

    I’ve asked you to follow the scientific method:
    1. Spot an oddity – done
    2. Get an idea – done
    3. Formulate a hypothesis – not yet achieved
    4. Test the hypothesis for non-coincidence – not yet achieved
    5. Present your research findings to the rest of the world – not yet achieved
    6. Defend your research findings by answering the questions expected from the rest of the world – not yet achieved

    You’ve gone from stage 1, where you’ve spotted an odd result, to stage 2, getting an idea. You then appear to fantasise that means you’ve achieved all the rest of the stages. Or maybe you imagine that you’ve achieved ultimate proof, definitively established cause and effect, and that you are now merely awaiting the Nobel?

    In reality, you’ve managed to get stuck at the stage many non-scientist or anti-science people end up at. And you’ve even started calling your idea a “study”.

    You haven’t been able to recognise logical extensions to your idea. (eg more than 10% impairment in sports performance would be immediately obvious over all sports, over all grades)

    You haven’t been able to recognise when your ideas are self-contradictory. (eg you tell us that more than 10% impairment in sports performance would only be evident in the very peak grades whilst telling us you’ve observed it in lower grades)

    You haven’t been able to answer valid questioning of your ideas.

    You have perceived valid questions of your ideas as a personal attack.

    It would appear that you are unaware that all scientists have to defend their findings and reasoning. I suggest that you look up what earning a PhD entails.

    All in all, soudhill, you’re not managing the scientific method very well.

    Perhaps you should follow my suggestion and attend high school classes in the subject?

    Like

  15. Stuartg: “Let us know who extended their fantasy to lower grade players by first mentioning lower grade rugby teams, such as Timaru and Petone.

    Let us know who first mentioned football teams in the lower UK leagues such as Birmingham City.”

    Timaru (Sounth Canterbury) was in Division 1 until fluoridation and producing lots of All Blacks.

    Petone was in the top Wellington competition (Jubilee Cup) until the Petone Tech closed and boys had to go to school in fluoridated Hutt.

    Birmingham CIty was winning against Manchester United and playing in the top competition (becoming Premier League). 8 years after fluoridation it was still playing in the top competition but not winning against Manchester United. So it was relegated.

    And Stuartg claims I am comparing top teams against lower ones. Only because they become lower some time after fluoridation.

    Like

  16. Stuartg wrote: “You haven’t been able to recognise logical extensions to your idea. (eg more than 10% impairment in sports performance would be immediately obvious over all sports, over all grades)”

    Unless the genetics and environment are different for the top players.

    Like

  17. Stuartg: “Then let us know who hasn’t been able to demonstrate significance in the tiny number of coincidences he’s been able to cherry pick among the thousands (millions?) of results available for him to peruse.”

    Going through your list to check here’s one I have come across that does not give a correlation: Aston Villa vs Blackburn Rovers. But the p value is so high using my system that the result cannot be trusted.

    While 47 clubs have competed since the inception of the Premier League in 1992, only six have won the title: Manchester United (13), Chelsea (4), Arsenal (3), Manchester City (2), Blackburn Rovers (1) and Leicester City (1). The current champions are Leicester City, who won the title in 2015–16. A small part of Leicester is fluoridated but I can’t find when it started.

    With 10% of England fluoridated shouldn’t 10% of the winning years tend to be from fluoridated areas?

    Like

  18. Stuartg: “and that you are now merely awaiting the Nobel?”

    No I know I make too many spelling mistakes.

    Like

  19. soundhill,

    Good spotting of unusual results. You’ve obviously had an idea. 💡

    Now it’s up to you to do something with that idea. Don’t just remain stuck there and complain no-one pays attention to your idea. Do something with the idea that’s worth someone paying attention to. Be scientific about it.

    First you’ve got to formulate a hypothesis.

    Then you need to test your hypothesis for non-coincidence by using fresh data. (Re-using data that was used to generate a hypothesis isn’t testing it). I’ve already pointed out how you can get such data from the ‘net.

    Then publish your observations, your data and your results.

    Then defend your data, thinking and results against scientists’ questioning.

    At that stage you may, or may not, have demonstrated a correlation between CWF and sports performance. You want it investigated further? Then you have to demonstrate non-coincidental correlation.

    Until you’ve done all of that, your appearance to the world remains that of an anti-fluoridationist living in a conspiracist fantasy world of your own imagination, one who is denying the many decades of epidemiological research that confirm the benefits and safety of optimally fluoridated water supplies.

    Like

  20. “Then you need to test your hypothesis for non-coincidence by using fresh data.”

    How about you testing your claim that anything which affects some teams must affect all,

    Like

  21. Ken if it makes a difference you may have used the maximum elevations rather than the average for the states. Colorado is the highest at 6,800 ft.

    Maximum elevation is only “explaining” a bit over 60% of average elevation

    Like

  22. soundhill,

    I make no claim other than that you appear to have spotted coincidence.

    You, however, claim to have observed detrimental effects (more than 10%) from CWF over multiple sports, multiple teams, multiple grades of sport, and multiple countries.

    But if you’ve changed your mind and are now no longer claiming there’s an effect on all teams, all sports, all grades, all countries, then you’ll have to allow for that when you produce your hypothesis before testing it. Your idea now begins to sound very close to the default of you having spotted coincidence.

    Personally, I think that you’ll be unnecessarily complicating your hypothesis if you start claiming the effect only occurs sometimes, some places, some sports, some grades: it would be like you, as an electrical technician, trying to find an rare intermittent fault on a circuit board – most of the time your testing would show nothing wrong with the board.

    Now, how about you providing some evidence of this effect (more than 10%) you claim to have observed?

    Bear in mind that repeating the coincidences that you’ve cherry picked to highlight your idea does not constitute providing evidence.

    Form your hypothesis from your cherry picked data, then use fresh data to test that hypothesis for non-coincidence. You could even pretend to be a scientist when you use the scientific method.

    Like

  23. Stuartg as with the “testing” of earthquake hypotheses which you related of, new types of foundations, it may take some time for new test circumstances to occur.

    Like

  24. Stuartg: “Bear in mind that repeating the coincidences that you’ve cherry picked to highlight your idea does not constitute providing evidence.”

    I relate them as a reminder then in discussion with you new ideas emerge that I have not seen before such as early this morning: “With 10% of England fluoridated shouldn’t 10% of the winning years tend to be from fluoridated areas?”

    I find it hard to do a correlation test over all the sorts of things I noticed, but maybe someone else can become interested.

    As I said I noticed unfluoridated Canterbury to be doing well. That formed my hypothesis then I started looking more widely. I don’t think myself to be seriously cherry picking.

    Like

  25. soundhill,

    “I find it hard to do a correlation test over all the sorts of things I noticed”

    Why can’t you do it? Your comments about other people’s​ statistical analyses imply that you are conversant with doing them yourself. Go ahead, no-one is stopping you. No data? Well, I’ve pointed out how to find sources for the new data that you need to do the testing. Why don’t you use them? Not doing the correlation testing suggests that you already know there is no correlation between sports performance and CWF.

    “But maybe someone else can become interested”

    Highly unlikely. After the multiple decades of epidemiological research into populations with optimally fluoridated water supplies that completely contradict your beliefs, scientists are aware that the prior probability that your observations are non-coincidental is close to zero. So why should they waste the time and energy?

    But maybe you could interest some of your anti-fluoride cronies in doing the correlation studies that are evidently beyond your abilities? Or maybe they also think you’ve noticed coincidence and so can’t be bothered?

    “That formed my hypothesis”

    No. Hypotheses are for testing. That you haven’t tested one in – how many years now? – shows that you don’t have a hypothesis. Just fantasies.

    **********

    You claim to have observed detrimental effects (more than 10%) from sports people drinking optimally fluoridated water, occurring over multiple sports, over multiple teams, over multiple grades of sport, and over multiple countries. You now say you are unable to provide evidence to back that claim.

    I think an applicable phrase is: put up, or shut up.

    Like

  26. Stuartg: “Highly unlikely. After the multiple decades of epidemiological research into populations with optimally fluoridated water supplies that completely contradict your beliefs,”

    I have asked for how many studies. Let’s make it easier, which of these ranges?:
    1-10
    11-100
    101-1,000
    1,001- 10,000

    Or better still give me some citations.

    Like

  27. Stuartg I wonder if you have ever seen cases of pellagra.

    Is ir showing again with poverty or idiosyncratic diets reappearing which were extremely uncommon when you went to medical school?

    Imagine yourself back in history you are seeing patients with varying degrees of dermatitis, diarrhoea, dementia, death.

    What sort of proof would you have wanted of a cause before treatment? Could you have encompassed the three or four symptoms into one disease or would you have had to separate them?

    How is correlation done when one cause can have a number of outcomes, as you note for my fluoridation/sport picture?

    And my fluoridation picture is further complicated by temperature/water consumption, modification of calcium/magnesium level and balance in water and the possibility for effects pre-conception to current with possibility for apparently delayed action.

    Like

  28. Stuartg you must have come across pellagra, possibly without knowing it and prescribing antidepressants, in patients on certain medications or even alcoholics.

    “The classical triad of pellagra is dermatitis, diarrhea and
    dementia. The symptoms do not have to appear in this order.
    Early symptoms of pellagra include lassitude, weakness,
    loss of appetite, mild digestive disturbances and psychiatric or emotional distress (anxiety, irritability and depression).”

    https://www.researchgate.net/profile/Vladimir_Hegyi/publication/227807440_Pellagra_Dermatitis_dementia_and_diarrhea/links/56c0990b08aeedba05647102.pdf

    You will say I am diverting the topic but I say you are trying to force an oversimplified approach to studying
    possible fluoridation effects.

    Like

  29. soundhill,

    As I said: put up, or shut up.

    Like

  30. Brian – thanks for pointing out my mistake with the figure where I use maximum rather than mean state elevations. I have checked it out and fortunately, the mistake is only with that graph (I pulled out some information at the last minute for the graph and must have been confused). Fortunately, I have used the correct elevation figures in all the statistical analysis, etc.

    I have also corrected the figures where I wrongly cited infant death rates as percent.

    I guess that comes from missing my daily power nap. 🙂

    Thanks again.

    Like

  31. Stuartg wrote: “As I said: put up, or shut up.”

    How about you stop shutting up and answer my questions which help elucidate about getting an agreed framework.

    Some of the time pellagra shows up as black tongue in creatures. But you can’t demand black tongue as a symptom before your treat with vitamin B3 rather than antidepressants like Prozac.

    Fluoridation could have one or more of several possible synptoms in sports teams from an area. The effects could be caused by
    1. quantity of water consumed by players
    2. change in the balance and concentration of calcium/magnesium when the water is treated.
    3. interaction with other issues such as low iodine and or COMT and MAOA gene variants which may affect drives.
    4. epigenetic effects – that gene switching on or off adaptation is inherited by offspring.
    5. effect of results of 1-4 above on secondary school players with a follow on result on the adult teams they may be at the top of in 8 years.
    6. possible other interactions or direct effects.
    So the sampling and correlation testing be not simple,

    Like

  32. Ken possibly to improve your analysis please note that the average dweller in a state may be in cities which are on average below or above the average elevation of the state.

    That if your elevation – preterm/morbidity/morality figures be correct they impact on the assumption in the Han/Tibetan blood oxygen saturation I cited, don’t they?

    Like

  33. It’s not my analysis, Brian, and I am not interested in digging deeper into searching for causes. I do not have such expertise.

    I was simply showing how that the information being promoted by the FFNZ is completely wrong.

    They are scaremongering.

    Just as they did with the offensive email to Green Party MP Julie Anne Genter:

    https://matterchatter.wordpress.com/2017/04/15/an-open-letter-on-fluoride-science-and-kindness/

    Like

  34. Ken though when research papers are claiming that altitude reduces oxygen saturation and causes trouble and your result claims altitude reduces trouble doesn’t that clash need to be sorted?

    Like

  35. Not by me.

    Are you attempting to divert again away from the lies being told by the FFNZ people?

    That is what the article is about.

    Like

  36. Ken your relationship with altitude could be a coincidence just as Stuartg claims about a result of mine even though it had an excellent p value.

    Like

  37. Of course it could be, possibly. Correlation is never causation.

    But we do know that the association with fluoridation is bogus, don’t we?

    Like

  38. Correlation is not causation. But the issue is that your p value of 0.002 means that there are 2 chances in a 100 that the correlation is only by chance before even thinking about causation.

    My correlation at 3:05pm https://openparachute.wordpress.com/2017/03/23/fluoride-coffee-and-activist-confusion/#comment-81572
    Results n=106, Rs=0.3041, t=3.26,
    p (two tailed) = 0.001506

    has a better p value than yours but Stuartg pointed out it could be coincidence.

    You must be careful not to be tending to get people to think into circular thinking when you say, “But we do know that the association with fluoridation is bogus, don’t we?” since you are basing the bogusity claim on the correlation aren’t you?

    Like

  39. No I am not – did you not understand the article.

    I am basing the bogus charge on the fact that there is no statistically significant correlation with percentage of fluoridation in a state. And when covariates are included the level of significance drops even further – through the floor.

    Fluoridation should be the last fact to consider.

    Like

  40. Ken your p value of 0.592 means that your calculation cannot be relied on for its assertion of low correlation.

    Like

  41. What do you mean, Brian? Do you interpret such a large p value as indicating a statistically significant association? Surely not!

    Like

  42. Ken the p value says whether the r value (correlation) obtained is “significant.”

    You haven’t given the r value you obtained. Whatever it be, high or low means little since the p value is high.

    To claim a low correlation you state it and p must be<0.05 or better.

    Al you have shown is that your calculation has not come up with a good enough p value to say anything.

    Like

  43. I am not the one saying anything – it is your mates in FFNZ claiming that fluoridation is associated with increased infant deaths.

    The extremely high p value shows they are wrong when elevation is included as a covariate. Even without including any covariates their association is still not statistically significant.

    There is absolutely no association between fluoridation and infant death – why try to claim there is?

    Like

  44. Ken: “The extremely high p value shows they are wrong when elevation is included as a covariate.”

    No it just shows that the data you have put into that sort of calculation is not sufficient to make any claim and the p value is telling you.

    The high p value means a correlation cannot be claimed using those figures and that method. But it cannot be ruled out.

    To rule it out the calculation needs to produce a very low r
    and p<0.05 or better.

    Like

  45. Ken you might try this, entering in say your r values you have for any relationships between say pre-term and fluoridation, pre-term and altitude, and altitude and fluoridation. I presume N will be 51 if you have data for every state. Vassarstats will do its sort of calculation and give t and two-tailed p values., holding each of the three variables controlled in turn.
    http://vassarstats.net/par.html

    Like

  46. By putting in larger values of N it is possible to see how many states would be necessary to give significant partial correlations when the non-partial correlations are low.

    Please do not confuse “significant” with notable – the magnitude of the correlation. “Significance” (the p value) only indicates whether the correlation which has been calculated can be trusted.

    Like

  47. soundhill,

    “Fluoridation could have one or more of several possible synptoms (sic) in sports teams from an area. The effects could be caused by”

    “could”? You’re not sure?

    The first thing you need to do then is to actually show if an effect actually exists. If the effect is “more than 10%”, as you say, then there should be plenty of evidence for you to cite.

    Why are you not able to cite evidence that the effect actually exists? Why are you unable to “put up”?

    The logical answer is that you can’t provide evidence for the simple reason that CWF has no effect on sports performance.

    Like

  48. Brian, you are welcome to try whatever you want in your efforts to retrieve a nasty role of fluoride. The data is freely available. But I am not going to waste my time.

    But there is clearly no association of preterm births with fluoridation extent – you can tell that just be eyeballing the data in the figure. Why bother attempting to revive the claim. It is clearly another lie from the FFNZ people. You are just attempting to cover that lie up with yet another diversion.

    Like

  49. Stuartg I have produced many observations putting fluoridation under suspicion. They are of differing scenarios. I am not all the way there yet as you aren’t when you prescribe mental health treatment without checking vitamin B3 status.

    Like

  50. soundhill,

    “the average dweller in a state may be in cities which are on average below or above the average elevation of the state.”

    Yep. Average is average.

    Like

  51. soundhill,

    “I have produced many observations putting fluoridation under suspicion”

    But you still haven’t produced any evidence that your observations are not coincidence. Until you do so the suspicion remains entirely your own fantasy.

    As I said: put up, or…

    Like

  52. soundhill,

    As Ken, along with many others, said: correlation is not causation.

    You haven’t yet demonstrated correlation. But you fantasise that you’ve demonstrated causation.

    You really should attend classes about the basics of science.

    Like

  53. soundhill,

    Perhaps someone else talking about the toxicity of fluoride may help?

    https://matterchatter.wordpress.com/2017/04/15/an-open-letter-on-fluoride-science-and-kindness/

    Like

  54. Ken perhaps you should also take into account induced abortion rate. On your preterm births chart the high points would be Mississippi, Louisianna, Alabama and West Virginnia. They are in the lower half of abortion rates. So might it be expected that less viable babies are being kept as long as possible?

    Those states are about half way long your altitude chart, making the line more horizontal than it would be if they were not at that point.

    Induced abortion rates are negatively correlated with altitude

    r -0.3638

    t -2.62

    N 45
    P two-tailed
    0.011946

    so presumably positively correlated with fluoridation extent, so going by averages these four states could be further along the chart increasing the slope.

    Like

  55. Have you not got the message, Brian?

    My article pointed out how FFNZ was using cherry picked data to lie about the relationship of infant deaths to fluoridation. I showed there is no statistically significant association – particularly when one possible covariate (which is related to fluoridation extent) is included. The weak and statistically non-significant relations they claimed is not due to fluoridation.

    As for preterm births – they advanced a conspiracy theory to imply measurements have found a statistically significant relationship but this information had been suppressed. I simply showed there is no statistically significant relationship, the data is readily available and they are lying about suppression.

    I have shown that yet again, FFNZ is lying and they a fiddling statistics. Yet again!

    My task is complete – I have absolutely no interest in doing any of the things you suggest. You are simply trying to divert attention away from the purpose of my article and what I have shown.

    That FFNZ is lying – yet again.

    Like

  56. And the top 3 of those states have African American percentages of MS 37.3 (2nd in US)
    LA 32.4 (3rd)
    AL 26.38 (7th)

    and when working with all subjects you still need to be fair comparing like with like, then the line would be steeper.

    Like

  57. One way of lying with statistics – in epidemiology – is to leave out categories who don’t support your case, as the CDC did with MMR, Black boys and ADHD. Or to include non-equivalent group members when it suits.

    Making a start to get equivalent groups I have left off Hawaii and the 10 states with about 20% or more African American population. (Actually 19.91% or above).

    Then correlating preterm births against fluoridation extent I calculate from Vassarstats

    number of states 40

    correlation 0.3628 or fluoridation “explaining” 13% of preterm births.

    t value 2.4

    P
    (two-tailed)
    0.021401

    Like

  58. soundhill,

    That last comment demonstrates two things you lack: knowledge of statistics and inability to question anti-fluoride propaganda.

    http://scienceblogs.com/insolence/2016/01/06/the-cdc-whistleblower-data-dump-redux-even-william-thompson-appears-not-to-believe-the-antivaccine-spin/

    But well done for trying to divert again.

    Like

  59. Stuartg so you make a big thing about where Thompson saved the computer files at the time of the document destruction: “All the associated MMR-Autism Study computer files have been retained on the Immunization Safety Office computer servers since the inception of the study and they continue to reside there today.”

    Like

  60. Stuartg:”you lack: knowledge of statistics”

    I think you to be referring to my race-based selection of subjects which then shows a result.

    What protocol would you use about comparing likes with likes?

    Like

  61. soundhill,

    I’d refer to a statistician, someone who knew more than me about the subject.

    I’m aware I don’t have the knowledge to comment.

    I only have two years of statistics at university so am much less qualified to talk on the subject than an electrical technician who hasn’t managed high school science.

    https://en.m.wikipedia.org/wiki/Dunning–Kruger_effect

    Like

  62. soundhill,

    “Thompson saved the computer files at the time of the document destruction”

    Haven’t you heard? Hooker’s incompetent “re-analysis” (subsequently retracted) was of Destefano et al’s already published data. He had no need to save data files already in the public arena.

    You continue to demonstrate that you are unable to question anti-fluoride doctrine.

    Like

  63. soundhill,

    I’m curious.

    Do you have an anti-fluoride catechism? One that you use for cut and paste? One that you’re not allowed to deviate from? Even when it’s been repeatedly proven wrong?

    You give that impression.

    Like

  64. Stuartg: “He had no need to save data files already in the public arena.”

    The files only remained on the computer because Thompson saved them. They are not exactly in the public arena now it is very hard to get permission to review them.

    Like

  65. Stuartg: “Do you have an anti-fluoride catechism? One that you use for cut and paste? One that you’re not allowed to deviate from?”

    No I think things out for myself. Another example 7 or 8 years ago
    http://tumeke.blogspot.co.nz/2009/03/someone-please-tell-national-that.html

    “At 27/4/09 3:47 pm, Blogger soundhill said…
    Ian Wishart wrote:
    “And memo to Bomber, the planet is currently cooling overall, not warming. Latest data, FWIW, shows Antarctica in particular has cooled by up to 0.21C since 1980.”

    I would expect Antarctica to cool with global warming. Deep ocean currents bring heat to calve icebergs which then travel to latitudes where they reflect away heat which was formerly being absorbed by seawater. ”

    I also sent emails to some scientists and the subject is turning up in Scientific American years later.

    https://www.scientificamerican.com/article/antarctica-rsquo-s-sleeping-ice-giant-could-wake-soon/?WT.mc_id=SA_DD_20170412

    Like

  66. Stuartg are you referring to this DeStefano paper?:
    http://www.vaccinesafetycouncilminnesota.org/Research/PediatricsDeStefanoMMRAtlanta113(2)259.pdf

    They specify male and female, and race but do not give race by sex.

    Like

  67. soundhill,

    An earlier comment of mine was:

    “I have produced many observations putting fluoridation under suspicion”

    But you still haven’t produced any evidence that your observations are not coincidence. Until you do so the suspicion remains entirely your own fantasy.

    As I said: put up, or…

    I notice that you still haven’t shown correlation, that you still haven’t shown that your observations are not coinidental.

    The obvious conclusion is that any relationship between CWF and sports performance that you noticed is pure coincidence.

    Unless you show correlation, that is…

    Like

  68. Coincidental, not coinidental.

    Like

  69. soundhill,

    “Stuartg are you referring to this DeStefano paper?:”

    No, I refer to the complete original paper.

    Like

  70. Stuartg my cite was a complete paper not just an abstract. If you think there to be something more “original” dividing Blacks by sex, then please cite.

    Like

  71. Stuartg I have shown several correlations.

    Like

  72. soundhill,

    You have not shown any correlations. What you have done is to mention some interesting coincidences.

    Nothing you have written or cited has demonstrated non-coincidence of those observations.

    I’ve pointed several ways that you can demonstrate non-coincidence of your observations, any of which could be easily used by an expert statistician.

    Once you demonstrate non-coincidence, then you have correlation.

    Unless you demonstrate correlation, all you have is coincidence.

    But then, expert statistician that you claim to be, you already know all of that.

    Failing to attempt demonstration of non-coincidence suggests that you already know there is no correlation and that your entire purpose is spreading FUD about CWF.

    Like

  73. soundhill,

    You want the original data from Destefano et al?

    Simple. Such data is never deleted as the authors may want to re-analyse when additional data becomes available. It’s still available from the CDC: https://www.cdc.gov/ncbddd/developmentaldisabilities/maddsp-data-sets.html

    Like

  74. Stuartg, from that how is anyone supposed to know what data sets are available?

    And Dr Hooker says Destefano was the one who had ordered the destruction of the data and took over from Thompson and presented fraudulent data in a talk to the Institute of Medicine which Thompson had been supposed to give.

    Like

  75. Stuartg a correlation can be by coincidence – chance. The associated p value tells of the chance the correlation is only the result of chance.

    At p=0.05 there is one chance in 20 that that correlation only happened by chance – coincidence. My p values were usually better. So my correlations are regarded as statistically significant.

    You may only claim I have not demonstrated a causal connection – the same with any set of correlations.

    The figures I gave for South Canterbury’s contribution to the All Blacks appears that I would get a significant negative correlation with fluoridation. That does not mean that I would get the same for other places say like Tauranga, Then I would look for other factors such as low South Canterbury iodine compared to Tauranga.

    Like

  76. soundhill,

    Small numbers mean unreliable statistics. Even a high school student of statistics could tell you that. But then, as an electrical technician who has never done science, your skills with statistics are obviously so much greater than someone who’s actually been taught statistics.

    “That does not mean that I would get the same for other places say like Tauranga” – and I’m encouraging you to do the same for places like Tauranga. Or New Plymouth. Or Southland. Or Marlborough. Or Northland… In fact, why don’t you include everywhere in the country so that you have significant numbers and can allow for confounding variables?

    I’ve pointed out how you can do the same for UK football teams. I’ve pointed out how you can check secondary level NZ teams. I’ve pointed out how you can check with different sports, using different grades, both sexes, all ages.

    …But you don’t do so.

    I suspect that you already know that you would demonstrate coincidence and non-correlation by checking other, larger, numbers.

    So, instead, you ignore the science and continue to use cherry-picked coincidences to spread FUD about CWF.

    Here’s a simple challenge for you, directly related to your “observations’ and “studies”:

    After over 70 years and many hundreds of millions of people enjoying the benefits of optimally fluoridated water, please cite one documented example of any physical harm to any person, or to any population, because they drank optimally fluoridated water, even for as much as a lifetime.

    Like

  77. Stuartg, if I do the same for Tauranga, then if the correlation for Tauranga be statistically significant, then it be worth comparing it to Timaru, if Timaru also be statistically significant.

    Like

  78. Stuartg perhaps an analogy will help.

    If I am on holiday Is it safer to drive a car on the right or left side of the road? Say I holiday in Australia of Japan and do OK on the left in a statistically significant fashion, then you are like someone who holidayed in the USA or China who claims driving on the left be not safe for them in a statistically significant fashion. You are like saying I ought take a larger sample over more countries.

    In that case there is the obvious difference where we know that the rules are different. Some countries drive on the left and some on the right. Taking the wider average does not help. If I holiday in several countries I should soon find out the two statistically significant patterns. Unless I am slow at learning and think there should be one way overall to be testing.

    If my thinking ability develops I may see there may be something to do with which side of the car the steering wheel is on. Or there may be arrows on the road.

    All that is so easy to see so why not for possible interactions of fluoridation with temperature, activity, amount of water drunk, in combination with other inputs, balances of minerals and sunlight/ vitamin D at various latitudes/altitudes/skin colours, other genetic types of person?

    Do you see that to leave out more principles and just ask for one overall tally for fluoridation be wrong, just as it be wrong to leave out knowledge of different driving rules in various countries?

    It is like you keep asking me for one example where driving on the left side of the road isn’t safe because it has been proven it works in the whole “British Commonwealth.” There may be better analogies.

    Like

  79. soundhill,

    Do go on…

    Everyone is noting how you are unable to provide any evidence that people or individuals have been harmed by drinking optimally fluoridated water in more than seventy years.

    Like

  80. soundhill,

    “perhaps an analogy will help”

    No. Evidence would help your argument. If you have any.

    Like

  81. Stuartg please define “optimally fluoridated water.”

    Are you going to give me the 1990 version or what?

    Why has the definition changed? If no harm then no need for change.

    Like

  82. soundhill,

    Unable to answer the question?

    You can use today’s definition, or that of 70 years ago, or any intervening, as long as you justify your choice.

    Everyone is still noting how you are unable to provide any evidence that populations or individuals have been harmed by drinking optimally fluoridated water for over seventy years, even over entire lifetimes.

    Like

  83. soundhill,

    Evidence would really help your argument.

    But you still haven’t produced any.

    Like

  84. soundhill,

    If you have difficulty in deciding what an optimal level of CWF is, you could always use the level from somewhere that democratically chose to continue CWF.

    Perhaps the level in Ken’s home, Hamilton?

    Like

  85. Stuartg: “As I said: put up, or…”

    Do you only accept overt symptoms?

    Like

  86. Stuartg do you have to see bandy legs in a child before you prescribe vitamin D? (By which time it be too late of course)

    Like

  87. Stuartg how large does a patient’s thyroid have to be before you are prepared to send them for a thyroid test? Don’t you think it be more fair to be awake to earlier signs?

    Like

  88. Stuartg: “I suspect that you already know that you would demonstrate coincidence and non-correlation by checking other, larger, numbers.”

    I have shown correlation figures and their associated p values which demonstrate statistical significance at much better than what may be accepted in many studies.

    What level of p value do you demand?

    Like

  89. Stuartg: “If you have difficulty in deciding what an optimal level of CWF is, you could always use the level from somewhere that democratically chose to continue CWF.

    Perhaps the level in Ken’s home, Hamilton?”

    In other words, not answering my question but proclaiming: “Obey me! Abracadabra you shall forget history.”

    Like

  90. It should be obvious that my references to overt symptoms, bandy legs and enlarged thyroid are follow-ons to earlier indications that could have been dealt with to prevent the eventual aggravated conditions. And that I maintain such matters as correlations between fluoridation and poorer community sports performance ought to be noted and acted upon before overt symptoms in individuals appear.

    Like

  91. soundhill,

    We can still see that you are unable to cite any evidence that individuals or populations have been harmed by drinking optimally fluoridated water supplies in the more than seventy years since CWF was started.

    Even when you are allowed to define “optimally” and “harm” yourself, provided you also justify your definition, you are still unable to cite evidence.

    What next? Do you need someone else to define the terms “water”, “drinking”, “fluoridated”, “seventy”, or “years” for you?

    Like

  92. soundhill,

    You may have shown “correlation figures​”, but you have not yet shown correlation or non-coincidence.

    But then, could we expect an electrical technician with no secondary education in science to understand that? Maybe not, especially when the Dunning-Kruger effect is so strong.

    Like

  93. Stuartg: “You may have shown “correlation figures​”, but you have not yet shown correlation or non-coincidence.”

    The likelihood of “coincidence,” or the calculated correlation value happening by chance is given by the p value. At p=0.05 it is a 1 in 20 chance, which is accepted in many studies as a statistically significant correlation.

    What p value limit do you require?

    Like

  94. Stuartg: “What next? Do you need someone else to define the terms “water”, “drinking”, “fluoridated”, “seventy”, or “years” for you?”

    The point is that “optimally fluoridated,” has changed from meaning a range: 0.7 – 1.2 mg/l to meaning only 0.7 mg/l in USA and NZ may be catching up.

    The story goes mild dental fluorosis is not an illness in fact it be good for you. But because it is increasing a little with better tooth brushing with fluoridated toothpaste the level put into water is going to be decreased. You weren’t being harmed it has been good for you but we’re going to decrease fluoridation level because of that increase in that good effect.

    Like

  95. “That good effect in which fluoride was damaging the cells which form tooth enamel causing the enamel to get develop with white opacities.”

    Like

  96. soundhill,

    …And the point is that even though you are able to specify your own level for optimal fluoridation, provided you justify that level, you still have been unable to cite an individual or a population that has been harmed by drinking such optimally fluoridated water, even for over seventy years consumption.

    Like

  97. And though fluoridation plus toothpaste was damaging the cells which form your tooth enamel especially in some people, and so causing the white opacities on the teeth, we don’t know if it be genetic and we don’t think it to be damaging anything else.

    Like

  98. Stuartg: “you still have been unable to cite an individual or a population that has been harmed by drinking such optimally fluoridated water”

    I have cited populations it is just that you have a mind set that illness has to be really bad before you accept it. Bad like too late to do anything much except arrange extra carers.

    Like

  99. soundhill,

    You have the option to define “harm”, as long as you are able to justify your definition.

    You have the option to define the level of fluoride at which “optimal fluoridation” occurs. Again, you have to be able to justify your definition.

    Now, given your ability to set those definitions, could you please explain to us why you are unable to cite any documentation of harm that has occurred to individuals or to populations from drinking optimally fluoridated water over at least the past seventy years?

    Of course, the default answer, the obvious answer, is that in more than seventy years of study epidemiologists have never found any harm to individuals or populations from drinking optimally fluoridated water.

    Your inability to cite evidence to the contrary is merely supporting the epidemiologist finding of no harm.

    Like

  100. Stuartg: “You have the option to define “harm”, as long as you are able to justify your definition.”

    Why not reduced community rugby performance?

    Like

  101. Stuartg: “You have the option to define the level of fluoride at which “optimal fluoridation” occurs. Again, you have to be able to justify your definition.”

    The accepted level for that community at that time.

    Like

  102. Stuartg: “could you please explain to us why you are unable to cite any documentation of harm”

    The discussion on this group could lead to eventual publication by me or others. Until then no citation of published work can be produced.

    You appear to try to be convincing readers that since nothing has been peer reviewed yet that it never can be. Funny, by your logic we would never see any new discoveries.

    Like

  103. soundhill,

    You believe that harm is/has been produced by CWF. At least, that’s what you tell us.

    So why is it that you cannot cite any documents that support your beliefs? Unless there is no evidence out there?

    After all, CWF has been used by millions of people for more than seven decades now. Hundreds, if not thousands, of epidemiologist have been studying those communities, searching for harm. Hundreds, if not thousands, of local and national governmental departments have been monitoring those communities – and most, if not all, of their monitoring data has been published online.

    And yet you are still unable to cite any published documents that support your beliefs.

    Maybe those unsubstantiated beliefs are better called fantasies?

    https://thelogicofscience.com/2017/04/23/science-matters-because-it-works/

    Like

  104. Stuartg: “You believe that harm is/has been produced by CWF. At least, that’s what you tell us.”

    Reduced community football performance for one thing

    “So why is it that you cannot cite any documents that support your beliefs? Unless there is no evidence out there?”

    Documents have to be written before they can be cited.

    “After all, CWF has been used by millions of people for more than seven decades now. Hundreds, if not thousands, of epidemiologist have been studying those communities, searching for harm. Hundreds, if not thousands, of local and national governmental departments have been monitoring those communities – and most, if not all, of their monitoring data has been published online.”

    How about 100 cites then? I doubrt they be what you imply.

    “And yet you are still unable to cite any published documents that support your beliefs.”

    It always has to start somewhere.

    “Maybe those unsubstantiated beliefs are better called fantasies?”
    Everything starts as a fantasy.

    You brought up the factor of trading of players between teams. I think it would work against my observations as players move from fluoridated to non-fluoridated areas or vice versa, not reinforce them. That will be something to eventually publish then people will be able to cite it.

    Like

  105. Stuartg cited: “Now, at this point, inevitably lots of people are going to get offended and respond with something to the effect of, “I’m not anti-science, but…I disagree with the way that science is being done, I think that massive corporations are buying off scientists, I have anecdotes that don’t match the science, scientists have been wrong in the past, scientists are close-minded, etc.,” but those aren’t valid responses and by using them you are standing in opposition to science, which makes you, by definition, anti-science.”

    Give a verification for that, please.

    You are not standing in opposition to science you are doing what science demands – challenge.

    Like

  106. Continuing, one thing we are led to believe is science, that is vaccines, is actually just a business venture.

    The vaccine manufacturers don’t even have need to work for safety since unlike most health areas they are absolved from any liability.

    Dr Hooker: “There is no inertia and there is no impetus for the CDC or the vaccine manufacturers to
    make safe vaccines. They know that vaccines are block-buster business. Many employees
    at the CDC end up in industry. Case in point Dr. Julie Gerberding, who was the director of
    the CDC from 2001 until 2008, took a very lucrative position as the head of the vaccine
    division in Merck in 2009. She was given stock options in the millions for that particular
    position so she overnight became a millionaire. There have been other employees that
    have gone on to lucrative positions. There’s actually a revolving door between the CDC
    and the vaccine industry.
    Dr. Thompson himself came from Merck. He worked at Merck before he worked at the
    CDC. Dr. Frank DiStefano who is the current head of the immunization safety office at
    the CDC actually left the CDC, went into industry and then came back to the CDC.”

    Like

  107. soudhill,

    OK, I understand what you’re saying. I suspect that so does everyone else.

    You fantasise that drinking optimally fluoridated water impairs sports performance. By about 10%, you said, which would reduce professional or Olympic grade athletes to about high school level.

    Absolutely no epidemiologists have ever noted that drop in performance, even though they’ve been actively searching for something like this for over seventy years. Neither have sports scientists – and they look for fractions of a percentage improvement or impairment.

    The truth is that you’ve observed some minor sporting coincidences, none of which have been duplicated elsewhere, and you now believe that with those coincidences you have proved harm occurs from CWF.

    Go ahead, write it up. Don’t forget to mention that you have no references to cite and that the entire thing depends on those coincidences you’ve observed.

    Even your anti-fluoride cronies will recognise it as fantasy.

    Like

  108. soundhill,

    All your comments up to date show you agree with the scientific maxim that the dose makes the poison; the higher the amount of fluoride ingested, the greater the degree of fluorosis on teeth and the greater degree of other effects.

    That maxim means that any effect of fluoride on sports performance is not going to be an on/off effect. It would be expected to be greater with increased fluoride intake, just as happens with other effects.

    The effect would be most evident in areas with high levels of fluoride in the water, exactly like the increased level of fluorosis in those areas.

    We can find such areas around the world: China, India, parts of the USA, many areas in Africa, areas in South America, areas of the Middle East, several of the ‘Stans, parts of Australia.

    How come no epidemiologists in those areas have noted a massive (“greater than 10%”, remember?) impairment of sporting performance from fluoride? Not even Chinese researchers, those that you have so frequently cited in the past for their dubious recognition of effects of high water fluoride, managed to observe the massive changes in sporting performance that you claim occur.

    Thousands of scientists, working worldwide, haven’t been able to detect an effect of CWF on sporting performance, in over seventy years of searching, even though they were actively searching for such things.

    Or what if the effect was not dose dependent, but instead it actually was an on/off effect, with optimally fluoridated water capable of triggering your claimed 10% impairment of sporting performance? 10% impairment would mean that NZ would not have any international level athletes from Auckland, Hamilton, Wellington, Dunedin… Indeed, there would not be any international level athletes from any areas of the world with CWF or higher water fluoride.

    We can readily see that’s not the case by looking at the sporting pages of most newspapers. Why look at p values when even The Press contradicts the idea of an on/off effect?

    p values (sigh!) You claim Birmingham City results to be statistically significant, p=0.05 for significance. https://en.m.wikipedia.org/wiki/List_of_football_clubs_in_England.
    Birmingham City is just one club in the top 27 leagues in England. Each league has about 20 clubs, some more. Each club has several teams. And you’ve noticed an effect on just one team from a list of well over 500 clubs? There should be many more teams with similar records, purely by coincidence. (Back of the envelope calculation: over 2,500 teams, so we should be able to identify similar records in over 120 teams)

    You have failed to identify a simple coincidence; instead you used that coincidence to reinforce your own beliefs, fantasising that you had identified cause and effect.

    Your logical reasoning is highly flawed, soundhill. Maybe if you had taken secondary school classes in science it would be better? Maybe you would understand statistics enough to recognise your lack of knowledge in the area? Maybe you would be able to identify coincidence?

    Like

  109. Stuartg: “Not even Chinese researchers, those that you have so frequently cited in the past for their dubious recognition of effects of high water fluoride, managed to observe the massive changes in sporting performance that you claim occur.”

    Tibet has to compete under the Chinese flag in the Olympics. Out of its over 3 million people it has only ever achieved one medal: a bronze in a women’s walk in 2012. Compare to New Zealand which achieved 18 for 4.5 million people last year. Tibetans drink a lot of brick tea which is high in fluoride.

    China with 1.7B people came 76th in the number of population per medal (70 total medals.)

    Like

Leave a Reply: please be polite to other commenters & no ad hominems.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s