a more recent study on vitamin D deficenccy and covid from the University of Chicago
This study has exactly the same problems: correlation does not imply causation.
THHuxleynew
Verified User
- Member since Jan 18th 2017
- Last Activity:
Posts by THHuxleynew
-
-
A study concluding vitamn D deficenccy and upper respiratory illness and it's relationship to covid 19
The evidence keeps building that the easiest and cheapest way out of this is as simple as supplementing vitamin D. The more I look into this the more it looks like vitamn D deficenccy and the effects on immune system has been overlooked by the medical community for many years
Fm1. I'd like you to consider carefully what this paper actually says, and what that means.
It identifies a strong positive correlation between low levels of vitamin D and COVID infection.
You, from your post, are interpreting this as low Vitamin D makes individual more vulnerable to COVID infection: perhaps by affecting the immune system.
That is of course possible.
But think on this. Suppose there were NO causal correlation between Vitamin D and COVID infection. Would you expect these results?
Yes.
There is are two obvious non-direct causal relationships between low Vitamin D and COVID:
Deprivation => low vit D. (Combination of bad diet, less attention to supplementation, less time exercising outside)
Deprivation => high COVID infection (Many factors the biggest of which is higher density housing, with less ability to socially isolate, also less resources to isolate, also jobs more likely to be high contact)
Second one is even simpler:
higher age => Low vit D (well known, absorption & transformation to calcifediol becomes less efficient)
higher age => Higher COVID infection (less strong immune system from many known age-related reasons)
Since both these factors are known, and would give a positive corelation, finding such a positive correlation does not add to evidence for the "vitamin D protects from COVID" hypothesis.
This is a very obvious example of why we need to be so cautious about observational data. The indirect causal relationships can be much more subtle than this. Attempts to get rid of them using statistics will go wrong unless strong factors like age are very carefully controlled. RB's most recent "positive for HCQ" evidence was a particularly bad example of this where the HCQ and control groups were highly mismatched in age distribution, and age was banded, rather than being given as a continuous parameter, with the bands covering risks varying by a factor of 3 (10 years). That makes errors due to uncompensated age differences inevitable, even if the correct stats is done to compensate for differences in age. (In fact they did not give any details of the analysis, I will revise this statement if given them and they prove better than it appears now).
This whole thread has put a lot of store (too much) by observational evidence without (usually) analysing whether correlations noted in results are properly causal, or something else.
I am still positive on Vitamin D for COVID. There is a lot of evidence - even though much of the observational evidnece is not real as above, there is also some better quality evidence. But, the bias against Vitamin D is because so many people look at the correlations without the necessary understanding that they may not be real. Those who do have that understanding see the arguments fro Vit D based on poor evidence and are not convinced.
Regards,
THH
-
Display More
Interesting research on mice.. via James Todaro,MD
Nasal priming with coronavirus protects against lethal lung infection
""In summary, we demonstrate that nasal exposure to coronavirus remotely primes lung immunity in the absence of direct lung infection, and that such priming provides effective protection against subsequent lethal viral infection in the lungs."
Of course well laid research on mice can oft gang aglay in humans..
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6124400/
Yes - it would be nice to see similar SARS-COV-2 work - but I guess they do not have controlled access to the virus so easily. Perhaps they do - if they can do vaccine challenge tests on rhesus monkeys they can surely test this.
-
Display More
Early HCQ results from analysis of ~5500 patient visits to ambulatory care..Saudi Arabia
"CONCLUSION: Early intervention with HCQ-based therapy in patients with mild to moderate symptoms at presentation is associated with lower adverse clinical outcomes among COVID-19 patients,
including hospital admissions, ICU admission, and/or death.
I looked at the proportions of patients in the >65 range. 10/1356 for HCQ (0.55%), 154/3724 SC (4.17%). Thus this group - the ones with vastly greater risk than the younger patients, which we would expect to contain most of the deaths and hospitalisations, has 10X more under SC than HCQ. The next at risk group (50-64) has 13.8% SC to 11.4% HCQ.
I had some other issues with the ways the stats were done (binning ages instead of using age as continuous parameter - though the details here are not made clear in the paper, if they has exact ages as well the could have used them as continuous variates in addition to the binning - however I think that unlikely. So, if this is the case, any results from this are unsafe because in these large bins the exact ages are critical to odds, and we know from the data that the age distributions are very different in the two cases that are compared).
I then looked at the comments - others had the same idea as me.
- Vincent Fleury • 4 days ago
Can you provide the distribution by age of the deaths, I can't find it in the paper. What I read is that there are 8 times more people in the stratum age>65yo, while the mortality is only 3 to 4 times higher. If mortality occurs only in the >65yo, then this work shows 1-that HCQ is not given to elderlies and 2-potentially that HCQ is actually harmful.
sunmaster14 Vincent Fleury • 2 days ago
Yeah, but it's not true that mortality only occurs in >=65yo. Worldwide, ICU admission and mortality for >=65yo is on the order of 25%. Applying that to the 154 old patients given standard care and the 10 old patients given HCQ would mean that 56 of the ICU admissions/deaths occurred for <65yo in the standard care pool, which is 1.6%, and 7 of the ICU admissions/deaths occurred for <65yo in the HCQ pool, which is 0.4%. You can obviously play around with the probability of >=65yo suffering morbidity/mortality with standard care, and get different relative results for HCQ on the <65yo subgroup, but using something so high that it implies everybody in ICU was >=65yo is obviously wrong.
Mansour Tobaiqy • 5 days ago • edited
We have shown that COVID19 patients on Hydroxychloroquine treatment had positive outcomes.
David Sbabo Mansour Tobaiqy • 3 days ago
No you didn't. You showed that patients with a much lower risk of dying are dying less than the ones with a higher risk. Congratulations.
- Vincent Fleury • 4 days ago
-
90% of the stuff you post is so wacky
the UK govt is (unsuccessfully) trying a Whack-a-Mole strategy for COVID.
I do however think it works on half-baked conspiracy theories. Wait till one emerges in a form clear enough to distinguish and "Whack" hit it with a few common-sense facts and a good dose of google.
-
a common medical term for this is PEP.. post exposure prophylaxis..
useful for some viral infections
Rhetoric is one thing, ignoring facts is another.
I referred to post-exposure, pre-symptom prophylaxis - what Boulware looked at.
You are calling post-symptom treatment prophylaxis which you can only do as a special case not particularly relevant here.
And my argument, which neither you nor the Wyttenbach have addressed, stands, regardless of a discussion about what should or should not be called prophylaxis.
May I respectfully suggest that you address the content of my arguments, instead of ignoring them and insulting me?
-
in other peoples personal details, trying to label them crazy or conspiracy theorists.
Navid, all Z needs to label crazygirl (5G, Gates causes COVID, germs are not real) here a conspiracy theorist is her weird self-avowed views. Much the easiest way to do it.
-
To many dilettantes do studies of all kinds. Just the first issue: People who get HCQ without having CoV-19 are obviously sick or persons at risk. To treat them as a cohort is more than fringe.
A pure sign of idiots behavior is to again post a study that just looks at HCQ, what nobody ever did recommend!
Why are proven dilettantes not able to distinguish prophylaxis after showing symptoms with prophylaxis in general ??
Wyttenbach: it only makes you look bad when you insult me, since your grasp of these matters, as judged by attention to detail, is worse even than that of the dilettante.
First thing. Prophylaxis after symptoms show is normally called treatment. Boulware, famously discussed here, showed results negative for days 3,4 after exposure, and possibly positive for days 1,2 after exposure (but not statistically significant unless you cherry-pick curve to fit). Sinvce symptoms are typically 5 days after exposure this is all about pre-symptom prophylaxis.
The study I linked, and that you have still not got to grips with (you should, it might give you new info) is pre-symptom, like Boulware, and also pre-exposure. It is thus the "best" possible case for prophylaxis. It includes the suggested idea that HCQ works as very early post-exposure treatment in days 1,2. Since HCQ takes time to build up a prophylatic dose before treatment is significant if continued for 6 months as here. we can argue details of dosing and pharmacokinetics if you like.
So if HCQ works as prophylaxis, which it might, this study should pick it up.
The cohort used is all of those having RA or SLE. These are "at risk" in the sense that they have a 19% higher mortality from COVID than the general population. That is a small amount: males are "at risk" relative to females by 100%. An extra 2 years of age puts somone this much more at risk.
But in any case what you need to understand (and avoid Wyttenfacting on) is that this study compared RA/SLE sufferers on HCQ (for pain relief) with RA/SLE sufferers NOT on HCQ (for pain relief). If HCQ has a significant prophylatic effect, as many here have claimed, it should show up in lower mortality for the with HCQ group relative to the without HCQ group.
As always, there are possible confounding factors, many of which have been compensated, but this evidence is a good deal stronger than that which has been posted positive on HCQ, and it is interesting.
A dilettante who tries to pay attention to details...
THH
-
Those pandemics killed 1 million worldwide and ~100,000 in the U.S. This one will probably kill 2 million worldwide and 300,000 in the U.S.
In addition (if you need more than a few 1,000,000 deaths) the after-effects for those seriously ill and recovered are a lot worse for this one than for Flu pandemics, because the cytokine storm (bradykinin blah blah...) stuff attacks other organs in the body and can leave them permanently damaged, not just lungs as for Flu.
-
to talk about the failure of psychiatry to even understand or treat the root causes of mental illness,
Agreed. We need to talk about over-consumption of conspiracy theories as possible root cause.
-
Fascinating Zeus, Navid has really out-done himself with that one. You know, Navid told you not to apply google search, so you have only yourself to blame. On the other hand, google search on speakers is so very much quicker than listening to long long youtube videos...
if he only took advice and started to pay attention to written stuff instead of youtube idiocy he'd have more context and be able to avoid the obvious fruitcakes?
Then maybe he'd have no-one left to quote and would need to post his own detailed arguments, with proper linked references. That would be fun to see.
-
I'd like those here who genuinely believe that HCQ is an effective COVID prophylactic treatment to review carefully this recent preprint, which approaches the issue from a different perspective and provides real hard evidence.
https://www.medrxiv.org/conten…101/2020.09.04.20187781v1
Basically, the take the whole UK population, divide all registered RA & SLE sufferers into those taking HCQ 6 months prior to COVID, and those not doing that. Look at mortality. Correct for everything else you can think of.
Results
Of 194,637 patients with RA or SLE, 30,569 (15.7%) received ≥2 prescriptions of hydroxychloroquine in the six months prior to 1 March 2020. Between 1 March 2020 and 13 July 2020, there were 547 COVID-19 deaths, 70 among hydroxychloroquine users. Estimated standardised cumulative COVID-19 mortality was 0.23% (95% CI 0.18-0.29) among users and 0.22% (95% CI 0.20-0.25) among non-users; an absolute difference of 0.008% (95% CI -0.051-0.066). After accounting for age, sex, ethnicity, use of other immunuosuppressives, and geographic region, no association with COVID-19 mortality was observed (HR 1.03, 95% CI 0.80-1.33). We found no evidence of interactions with age or other immunosuppressives. Quantitative bias analyses indicated observed associations were robust to missing information regarding additional biologic treatments for rheumatological disease. We observed similar associations with the negative control outcome of non-COVID-19 mortality. Interpretation. We found no evidence of a difference in COVID-19 mortality among patients who received hydroxychloroquine for treatment of rheumatological disease prior to the COVID-19 outbreak in England.
Methodology
We used Cox regression models with days since the index date as the timescale to estimate hazard
ratios (HRs) and 95% confidence intervals (CIs) for the association between regular
hydroxychloroquine use and COVID-19 mortality. The competing risk of death from causes other
than COVID-19 was accounted for by censoring non COVID-19 deaths; our analysis therefore
estimated cause-specific hazards.19 We sequentially adjusted for sex and age using restricted cubic
splines; for the minimal adjustment set informed by the DAG; and finally extended for all extracted
covariates listed above. Models were stratified by an indicator variable denoting patient population
(i.e., RA or SLE) and geographic region.Comments
My naive "is it OK" test is sort of alright here. they are not using (what seems to me good) logistic multivariate regression. Instead they are using cubic splines to deal with non-linear age in a cox regression. I don't properly understand Cox regressions, but I do understand cubic splines.
What I like:
Whole population method => large sample size and less possibility of selection effects
Differential effect can be detected (similar principle to RCT, though does not remove confounders that correlate with HCQ suitability)
Would definitely pick up any prophylactic effect of HCQ if this exists at the levels given to RA sufferers - so this is new (negative) info.
This is UK: no likelihood of any political bias.
What I don't like:
I'm not knowledgable enough to understand cox regressions (that just means I cannot evaluate this so well, does not mean it is not good quality)
This is RA sufferers, so there might be something going on their that changes things? No evidence for this however.
Always possible there are errors here due to unconsidered confounding variables (as with all non-RCT data).
Why they needed to adjust for age, sex etc:
Hydroxychloroquine users were younger (median: 63 years for users, 66 years for nonusers) and more likely to be female (76% women for users; 70% for non-users); other demographic
characteristics between exposure groups were broadly similar.So unadjusted that is a significant (50%) worse result for the non-HCQ group just based on age and sex, two factors where we well understand the relationship to mortality.
Comment on the preprint:
Hello, I could not understand how the estimated standardized cumulative death rates are 0.23% for HCQ group and 0.22% for NONHCQ group.
Deaths NONHCQ 477, sample 164.068 = 0,29%.
Deaths HCQ 70, sample 30.569 = 0,228%.
I know it is standardized, but shouldn't it be around the average number?
The standardisation get rid of the +50% non-HCQ seen in the raw data. Before anyone argues the raw data shows that HCQ works: really? You think age and sex don't affect mortality?
-
hydroxychloroquine triple therapy 1700 patients... zero deaths
like i said there were the kids in those age groups we did not prescribe it
because it's not indicated for kids right
RB - did you have some intent in posting (without link) this text? Claims of < 0.1% observed mortality from COVID are not surprising if the demographic of those treated is young: your text gives no indication of median age, except this offhand reference to kids. And in this case I'm wondering even whether the "patients" have been PCR tested?
It would be more useful as evidence with more details.
Approx rule of thumb: Age 60 - 1% Age 40 0.1% Age 20 0.01% (these are +/- quite a lot, but give an idea of the way mortality varies with age).
-
you may think you have recovered only to experience worse symptoms like kidney failure or neurological disorders later on
I believe there is no evidence for this interpretation.
The long-lasting issues are explained by the bradykinin hypothesis, which predicts all blood vessels to leak in severe COVID therefore damaging heart, brain, kidneys, etc and generating the wide variety of side effects shown. No hidden virus needed. There is now I think quite a lot of supporting evidence for this.
-
Why not [Gates] figure out a way with all your billiionaire friends to feed them since someone malnourished doesn't have the immune system to fight a virus very well.
These people are criminals.
So, how much money have you given to this cause? Better than $776M?
-
Only a truly diabolical figure like Gates
Navid - you are not helping your case for being considered seriously when you liken Gates to the devil.
These people are criminals
Or claim that Fauci is a criminal. I'm not entirely sure that is even legal. Certainly not fair comment. It makes no sense to me.
-
Display More
This sounds like reasonable evidence to me:
(Page 6) "Comparison with other treatments
Overall mortality rate of patients treated according to IDEA protocol was 0.59 % (1 death in 167 treated cases).As a comparison, estimated overall mortality rate in Argentina is approximately 2.1 % (official data by September 2nd, 2020).
Regarding moderate to severe cases, i.e. patients needing hospitalization, only 1 patient out of 32 receiving IDEA treatment
died (3.1 %), whereas mortality rate published in articles from Spain, Italy and Spain is ca. 25 %. Moreover, a group of 12
patients were hospitalized in Eurnekian hospital in the same period but did not receive IDEA treatment. Three of them died,
thus presenting a mortality rate of 25 %, i.e. significantly higher than that of those receiving IDEA treatment."
Also note that the moderate-to-severe average age was nearly 60, and that they were using oxygen or ventilated.Let's leave the doctor-patient team decide the best course, but who would want to be on the
placebo arm of an RCT? However, if they do, then they should be allowed.
Allowing people to choose placebo arm of RCT would invalidate the results. Completely. Allowing them to choose 100% novel treatment or RCT or best standard care would be OK.
There are many many observational trials which do multivariate logistic regression to separate out the wanted effect from other parameters: comorbidities, age,sex. At least you need to detail all of those parameters, and then see how your mortalities scale with equivalents.
They are not doing that. It is as though people crediting these results think COVID death rates vary only slightly with other parameters.
There is no evidence here this treatment does not deliver good results. And no evidence it does. No discussion of statistical significance either. For the death in hospital figures they have only 32 of which one dies, that is (probably? They would need to do the calculation) not a statistically significant difference when compared with 12 and 3 died, expecially when it is an ad hoc comparison not previously registered. (There are always many such ad hoc comparisons that can be made after you have the data, some of which by chance will be very positive).
Similarly, with only 1 death they would need to work out the statistical significance of the differnece from the national mortality - and in any case that for obvious reasons is a very poor comparator.
Small individual trials (even if RCT) have a tough time getting good data out, you need a lot of patients if your measure is mortality and 2% die. If your measure is determined by doctors (severity level, or hospitalisation, etc) you need rigorous specified protocols for who you do this to, and even then without masking there is likely bias. Having done that, you then need to control properly for other variables. These people have both small trial and no control of other variables against them. Also you might ask how many such small local trials there are - of which some will deliver positive results by chance and be written up as here. Would you ever write up a negative trial that is unregistered?
That is why large systematic regsitered methodology determined prior to trial trials like RECOVERY are useful - they have the statistical power (when COVID rates are high) to deliver answers to these questions.
THH
-
Display More
Successful ivermectin results from Argentina -
SAFETY AND EFFICACY OF THE COMBINED USE OF IVERMECTIN,
DEXAMETHASONE, ENOXAPARIN AND ASPIRIN AGAINST COVID 19
https://www.medrxiv.org/conten…09.10.20191619v1.full.pdf
Recent Zelenko interview - claims several successful treatments have now been
identified (HCQ, ivermectin, ...) but politics obstructs their use --
Zev Zelenko MD on Sep 11
External Content www.youtube.comContent embedded from external sources will not be displayed without your consent.Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.None of the patient presenting mild symptoms needed to be hospitalized. Only one patient died
(0.59 % of all included patients vs. 2.1 % overall mortality for the disease in Argentina today; 3.1
% of hospitalized patients vs. 26.8 % mortality in published data). IDEA protocol appears to be a
useful alternative to prevent disease progression of COVID-19 when applied to mild cases and to
decrease mortality in patients at all stages of the disease with a favorable risk-benefit ratio.Lou how do these guys know their preferred treatments are effective? Seems to me without some comparison that is unbiassed (= RCT) they cannot have much evidence? I mean, almost all (99%) of early COVID patients recover. That percentage changes by 10X (e.g. 0.3% to 3%) for a change in median age from 30 to 50 etc. many other effects change it too. I'm sure you know that mortality figures vary wildly according to how many people getting milder versions of the disease are counted, how many are treated in hospital (for the hospital mortality figures), etc.
Comparing local figures - esp when dealing with mild COVID cases - with national ones can mean just that you are treating more people - which you will by definition be doing if you are looking at early-stage cases.
it is such a shame that when there are possible effective treatments the guys doing this do not provide evidence they work.
-
THH I'm glad you brought that up. It would seem that you haven't read the studies I've posted concerning vitamin D deficenccy. Otherwise you would have read that vitamin D plays a major role in motabolism.
FM1 - nutrition plays a major role in metabolism! Look back at my previous posts re vit D. And the perils of assumng you understand what it does or does not do based on the current not easy to interpret evidence.
Still worth making sure you are not Vit D deficient.
-
Dr. Fucci vitamin D dosage vs what US government recommends by age group
Anything to do with nutrition (Vitamins count) and the evidence is so weak that experts can be genuinely divergent on what is right, and innocently wrong. Look at all the controversy over whether fat reduction in diet is good or bad.
That such can start conspiracy theories is a pity.
THH
