Changes

*Impossible to collect the data in the way the researchers claimed without violating privacy laws<ref name =Guardian>[https://www.theguardian.com/science/2020/may/28/questions-raised-over-hydroxychloroquine-study-which-caused-who-to-halt-trials-for-covid-19 Questions raised over hydroxychloroquine study which caused WHO to halt trials for Covid-19] by Melissa Davey (27 May 2020 22.27 EDT) ''The Guardian''</ref><ref name =arkancide/>

*Impossible to collect the data in the way the researchers claimed without violating privacy laws<ref name =Guardian>[https://www.theguardian.com/science/2020/may/28/questions-raised-over-hydroxychloroquine-study-which-caused-who-to-halt-trials-for-covid-19 Questions raised over hydroxychloroquine study which caused WHO to halt trials for Covid-19] by Melissa Davey (27 May 2020 22.27 EDT) ''The Guardian''</ref><ref name =arkancide/>

*Not enough authors to conduct a study of its size<ref name =Guardian/>

*Not enough authors to conduct a study of its size<ref name =Guardian/>

*Too many Australian deaths... it's data claimed to have recorded 73 Australian deaths by April 21 but the number of COVID-19 deaths in Australia didn't reach 73 until April 23.<ref name =Guardian/>}}

*Too many Australian deaths... it's data claimed to have recorded 73 Australian deaths by April 21 but the number of COVID-19 deaths in Australia didn't reach 73 until April 23.<ref name =Guardian/>

*Not adhering to standard practices in the machine learning and statistics community<ref name=JamesWatson/>

*Not adhering to standard practices in the machine learning and statistics community<ref name=JamesWatson/>

*Implausible ratios of chloroquine to hydroxychloroquine use<ref name=JamesWatson/>

*Implausible ratios of chloroquine to hydroxychloroquine use<ref name=JamesWatson/>

==Observational Study of Hydroxychloroquine in Hospitalized Patients with Covid-19==

==Observational Study of Hydroxychloroquine in Hospitalized Patients with Covid-19==

This study authored by Geleris is a retrospective study commonly cited in the media that analyzed patient data from the New York–Presbyterian Hospital (NYP)–Columbia University Irving Medical Center (CUIMC).  The following are inherent problems in this study.

This study authored by Joshua Geleris is a retrospective study commonly cited in the media that analyzed patient data from the New York–Presbyterian Hospital (NYP)–Columbia University Irving Medical Center (CUIMC).  The following are inherent problems in this study.

*[[Hydroxychloroquine]] was given late.<ref name=Geleris>[https://www.nejm.org/doi/10.1056/NEJMoa2012410?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed Observational Study of Hydroxychloroquine in Hospitalized Patients with Covid-19] by Joshua Geleris et.al (June 18, 2020) ''The New England Journal of Medicine''</ref>}

*[[Hydroxychloroquine]] was given late.<ref name=Geleris>[https://www.nejm.org/doi/10.1056/NEJMoa2012410?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed Observational Study of Hydroxychloroquine in Hospitalized Patients with Covid-19] by Joshua Geleris et.al (June 18, 2020) ''The New England Journal of Medicine''</ref>}

*The treatment with [[hydroxychloroquine]] varied depending on which doctor was seen.<ref name=Geleris/>

*The treatment with [[hydroxychloroquine]] varied depending on which doctor was seen.<ref name=Geleris/>

==Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in New York State==

==Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in New York State==

This study authored by Rosenberg is another retrospective study commonly cited by the media.  It analyzed the data from 25 different hospitals in the New York area.  The following were problems with this study.

This study authored by Eli Rosenberg is another retrospective study commonly cited by the media.  It analyzed the data from 25 different hospitals in the New York area.  The following were problems with this study.

*The treatment with [[hydroxychloroquine]] varied depending on which doctor was seen.<ref name=Rosenburg>[https://pubmed.ncbi.nlm.nih.gov/32392282/ Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in New York State] by Eli S Rosenberg et. al. (2020 May 11) ''JAMA''</ref>   

*The treatment with [[hydroxychloroquine]] varied depending on which doctor was seen.<ref name=Rosenburg>[https://pubmed.ncbi.nlm.nih.gov/32392282/ Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in New York State] by Eli S Rosenberg et. al. (2020 May 11) ''JAMA''</ref>   

*It was given to sicker patients preferentially (when compared to the control group).

*It was given to sicker patients preferentially (when compared to the control group).

==Hydroxychloroquine in Nonhospitalized Adults With Early COVID-19: A Randomized Trial==

==Hydroxychloroquine in Nonhospitalized Adults With Early COVID-19: A Randomized Trial==

This is the second of the two studies that come from the University of Minnesota.  The following are its flaws:  

This is the second of the two studies that come from the University of Minnesota.  The following are its flaws:  

*Only 58% of patients tested positive for COVID-19.

*Only 58% of patients tested positive for COVID-19.<ref name=Minn/>

*The data was self-reported  

*The data was self-reported<ref name=Minn/>

*The treatment was delayed as it was being delivered through the mail.

*The treatment was delayed as it was being delivered through the mail.<ref name=Minn/>

*The sample size was not large enough to achieve statistical significance.

*The sample size was not large enough to achieve statistical significance.<ref name=Minn/>

*The assessment "overall symptom severity" was subjective and not rigorous.<ref>[https://pubmed.ncbi.nlm.nih.gov/32673060/ Hydroxychloroquine in Nonhospitalized Adults With Early COVID-19: A Randomized Trial] by Caleb P Skipper et. al. (2020 Jul 16) ''Annals of Internal Medicine''</ref>

*The assessment "overall symptom severity" was subjective and not rigorous.<ref name=Minn>[https://pubmed.ncbi.nlm.nih.gov/32673060/ Hydroxychloroquine in Nonhospitalized Adults With Early COVID-19: A Randomized Trial] by Caleb P Skipper et. al. (2020 Jul 16) ''Annals of Internal Medicine''</ref>

During the course of this study, the researchers discovered that the low event rate of hospitalization and deaths among the participants meant that the trial would have required 6000 participants in order to achieve meaningful results.  Consider the following example in order to understand the nature of the problem.  Lets say that 1 out of 100 people infected with SARS CoV-2 will need to be hospitalized, the study has 400 participants evenly divided between the control and placebo groups, and hydroxychloroquine is known to be 50% effective.  The study is executed and 2 people in the control group are hospitalized whereas only 1 person in the treatment group was hospitalized.  One could argue that confirms the hypothesis, but anyone familiar with statistics would tell you that that is not correct.  The difference between 1 and 2 people being hospitalized could have easily happened due to chance and thus the difference is not statistically significant.  The only way to achieve statistical significance would be to increase the sample size, in this example to 6000 people, which would result in 30 people being hospitalized in the control group and 15 being hospitalized in patients treated with hydroxychloroquine.  That wasn't done and so the results are meaningless.  What was done was analyzing whether or not one self-reported having a cough after 14 days in both the placebo and treated groups and then trying to use that to determine if the treatment was successful or not.  That, to be perfectly frank, is not science (unless one was trying to evaluate the effectiveness of cough medicine).           

During the course of this study, the researchers discovered that the low event rate of hospitalization and deaths among the participants meant that the trial would have required 6000 participants in order to achieve meaningful results and the study only had 491.  Consider the following example in order to understand the nature of the problem.  Lets say that 1 out of 100 people infected with SARS CoV-2 will need to be hospitalized, the study has 400 participants evenly divided between the control and placebo groups, and hydroxychloroquine is known to be 50% effective.  The study is executed and 2 people in the control group were hospitalized whereas only 1 person in the treatment group was hospitalized.  One could argue that this confirms the hypothesis, but anyone familiar with statistics would tell you that that is not correct.  The difference between 1 and 2 people being hospitalized could have easily happened due to chance and thus the difference is not statistically significant.  The only way to achieve statistical significance would be to increase the sample size, in this example to 6000 people, which would result in 30 people being hospitalized in the control group and 15 being hospitalized in patients treated with hydroxychloroquine.  That wasn't done and so the results are meaningless.  What was done was analyzing whether or not one self-reported having a cough after 14 days in both the placebo and treated groups and then trying to use that to determine if the treatment was successful or not.<ref name=Minn/> That, to be perfectly frank, is not science (unless one was trying to evaluate the effectiveness of cough medicine... even then one would need a larger sample size).