@stuartbuck1 @StevePhillipsMD Stuart, this appears to be a well run study but there are concerns:
1) This deals with morbidity amongst those confirmed to have contracted the disease, not whether it prevented infection. Depending on criteria for testing in the UK this could be a very large gap.
1/
@stuartbuck1 @StevePhillipsMD It certainly does not robustly examine prevention of infection. It does examine mortality, which is important and may be correlated with infection.
2/
@stuartbuck1 @StevePhillipsMD 2) The criteria for use of HCQ was two or more prescriptions over a 6 month period but there is no data on dosage or current status of treatment at the time of symptom onset. This would not allow specific differentiation of effects of treatment being current at time of onset.
3/
@stuartbuck1 @StevePhillipsMD 3) These populations are users of DMARDS which may be a mortality confounder. This may be more important in light of no data on infection rates. It is a shame they were unable to tease this out of the data. But there were insufficient controls on exposure to draw conclusions.
4/
@stuartbuck1 @StevePhillipsMD 4) They are using a limited protocol without examination of other SOC and prior status factors such as vitamin D, Zinc, steroid use, Bromhexine, use of antibiotics or other protocols.
5) There is some concern about the age and sex adjustment process that bears further study.
5/
@stuartbuck1 @StevePhillipsMD This is a good study. The authors properly caveated their own findings. It provided further evidence HCQ is safe. The authors themselves limit the scope of their conclusions. I am a strong proponent of correlation being used to point out potential relationships to study.
6/
@stuartbuck1 @StevePhillipsMD It would have been much stronger had they been able to include other potential confounders, teased apart infection risk differences, and discussed the sex and age adjustment more fully(since it turns positive effect into no effect).
/end

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Harry Haury

Harry Haury Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @letsthinkdeeply

5 Sep
@DanCrenshawTX @gummibear737 There are a number of things you need to know about CIVID19 that are not talked about a lot that are causing a lot of confusion right now:
1) Young people below 45 without identified pre-existing conditions very rarely have severe disease or die.
1/
2) Many antibody tests also sometimes referred to as AB tests have a significant false positive rate due to reactive antibodies to other coronaviruses. There is controversy using them where the error rate is a meaningful percentage of the infection rate.
2/
Antibodies can also be positive when the disease has passed, for this reason we still do not have a good handle on the number of asymptomatic cases.
3) PCR, polymerase chain reaction, tests also have an issue in that they can find dead virus segments.
3/
Read 6 tweets
30 Aug
@littllemel Please be careful with this. I believe many things about the COVID19 crisis are purely political. But anything that infers only 9,210 people died because of C19 is simply wrong. There are big issues that need examination! The pandemic crisis has always been misunderstood.
1/
@littllemel The real issues:
1) over amplified PCR tests falsely indicating COVID19,
2) misrepresentation of comorbidity role in death, current guidelines label deaths as involving C19 that don’t really. These are certainly in the tens of thousands.
2/
@littllemel 3) comparisons between countries in measurement of the epidemic are completely irrelevant when the statistics use different methods.
4) the real scandal is we could have protected our vulnerable, saved trillions of dollars and kept our lives going.
3/
Read 6 tweets
30 Aug
@Covid19Crusher It is an issue I have discussed frequently. There is a common confusion amongst doctors, politicians and particularly the press concerning what these trials tell us. Several of the RCTs used to claim Hydroxychloroquine does not work actually suggest it does.
1/
@Covid19Crusher In order to try to avoid Alpha error, the confidence limits are set quite “tight”. To understand the effect refer to the standard Alpha versus Beta error graph.
2/
@Covid19Crusher Trying to “prove” if HCQ works the researchers require the confidence level to be as high as reasonable. But statistics are not exact. The higher the confidence limits, the larger your Beta error, in this case concluding Hydroxychloroquine does not work when in fact it does.
3/
Read 10 tweets
30 Aug
@TDSisdangerous @StevePhillipsMD The statistics and principals are a bit complicated but observational studies with retrospective analysis simply track a large number of patients using one or more treatment protocols. They tend to be larger because of the way the analysis examines many factors.
1/
@TDSisdangerous @StevePhillipsMD For instance, knowing about comorbidities, age, patient status, blood chemistry, and when treatment began seem to impact the outcomes they would track many of these data points for each patient.
2/
@TDSisdangerous @StevePhillipsMD A well designed observational study will try to track all the relevant factors they can, particularly ones shown previously to be important. As more and more of these studies are run the scientists know more about what they should track.
3/
Read 18 tweets
28 Aug
@abirballan @gummibear737 I think you would find that Gummi Bear 737 means that with a sense of caution many conceded 15 days to lower the curve was a reasonable precaution given the lack of knowledge we had. Obviously the lack of adherence and high degree of mobility made the effort mostly meaningless.1/
@abirballan @gummibear737 The biggest surprise was that initial epidemic curves, the lack of accurate tests, and as yet undiscovered cross reactive immunity had disguised the fact herd immunity appeared to occur around 15-20% instead of around 60-70%.2/
@abirballan @gummibear737 The IFR was not a surprise and was roughly in line with estimates. There is a group of us that have insisted we need an honest dialogue, in the initial uncertainty and confusion decisions are often wrong.3/
Read 9 tweets
22 Aug
@gummibear737 It is another manipulation of the Hydroxychloroquine debate. For reasons not entirely clear, the FDA has opened its FAERs database to monitor Adverse Medical Events that may have included Hydroxychloroquine. 1/
@gummibear737 This database is a purely observational reporting system used by the FDA when they are "concerned" about drug safety. We are intrigued by this step, and alarmed that a new front has been opened in the COVID19 and HCQ debate. 2/
@gummibear737 Why? It is simple, many RCTs have been run my the NIH on very sick patients late in the disease process, they all clearly show that there are almost no Hydroxychloroquine safety concerns where the patients in severe distress are being monitored and screened for cardiac events. 3/
Read 10 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!