On 2021-04-23 17:14:04, user Sam Smith wrote:

Don't worry about it.<br /> An analysis of data from the U.K. estimated that, beginning 14 days after the second dose, the Pfizer vaccine was 85% effective in preventing symptomatic COVID-19 in people 80 years and older compared to those unvaccinated (Bernal, medRxiv 2021 — preprint).<br /> https://www.medrxiv.org/con...

More studies:<br /> https://www.consumerlab.com...

On 2021-09-13 21:04:36, user Thomas Mohr wrote:

If I review a paper, I start with the methods section. So let us do this:

Quote: "We searched the Vaccine Adverse Event Reporting System (VAERS) data for <br /> females and males ages 12-17 in reports processed from 1/1/2021 through <br /> 6/18/2021 with diagnoses of “myocarditis,” “pericarditis,” <br /> “myopericarditis” or “chest pain” in the symptom notes and required the <br /> term “troponin” in the laboratory data. We defined a CAE using the CDC <br /> working case definition for a probable case."

As others, and VAERs, have pointed out, the VAERs database can not be used to do this.

Quote: "Cases and hospitalizations with an unknown dose number were assigned to dose 1 or dose 2 in the same proportion as the known doses: 15% occurred following dose 1 and 85% occurred following dose 2."

Nope. That is not how it works. Incomplete data have to be excluded.

Where is the proper prior probability of myocarditis?

At this point latest I would stop reviewing and return the verdict: reject without encouragement to resubmit.

On 2021-08-04 20:20:55, user BiotechObserver wrote:

For those downvoting me, your ignorance is showing. It is considered unethical to continue providing trial participants with a placebo once the intervention has proven beneficial. The trial participants must then be offered the intervention. Haven't any of you heard of the Tuskegee experiment? Try looking up why that is considered unethical.

On 2021-10-03 02:28:34, user OBS wrote:

How come this preprint (and the very recent publication of this in NEJM) both say 15 deaths vaccine vs. 14 deaths placebo, but the FDA briefing document for the booster shot (which summarizes the safety of the primary 2-dose series, see page 7), says 21 deaths vaccine vs. 17 deaths placebo?

https://www.fda.gov/media/1...

21 vs. 17 doesn't seem to be an update of the 15 vs. 14 result, since the booster FDA briefing document specifies March 13, 2021 as the data cutoff date corresponding to the 21 vs. 17 result, and that is the exact same cutoff date mentioned in this preprint / NEJM article. So why the discrepancy- what is going on here?

On 2021-09-24 20:53:59, user Kenny wrote:

Very surprised that there is no evidence of lowering total mortality. Essentially there is absolutely no direct evidence that vaccine saves life.

On 2021-08-22 00:51:09, user Anthony Loera wrote:

Supplementary data shows deaths between Vaccine people and controls are close to even<br /> https://www.medrxiv.org/con...

On 2021-07-31 09:02:29, user Jeff Jaffers wrote:

Where is Table S4?

On 2021-08-17 06:41:30, user Jeff Brender wrote:

https://www.cdc.gov/mmwr/vo...<br /> Kentucky residents who were not vaccinated had 2.34 times the odds of reinfection compared with those who were fully vaccinated (odds ratio [OR] = 2.34; 95% [CI] = 1.58–3.47) (likely the Alpha strain based on the date)

On 2022-01-06 13:41:46, user Kenneth Morton wrote:

Such a shame that with such a complete dataset, the unvaccinated results have purposely been contaminated with the 'single jabbed' and have also not been split between those previously uninfected and those who have been infected and recovered previously.

On 2020-07-19 05:03:49, user HarryDeedra Hodges wrote:

The tweet storm is quite odd. Seems like a mini-army, particularly in Spanish moving the story along. Others seems quite happy to spread the negative results. The patients seem evenly matched, but nearly all needed O2 supports implying quite ill, more HCQ patients were > 70 than the control. Dosage was 1600mg loading, 400 mg daily for 9 days, no mention of zinc. No adverse effects noted. The dosage is only slightly higher than the evolving standard which includes zinc. The study shows HCQ without zinc not useful in critically ill patients.

On 2020-07-28 11:19:04, user Joe wrote:

I wonder if the key sentence in this isn’t in the last full paragraph: “The findings indicate that (HCQ) is not an effective treatment for hospitalized patients with COVID-19 but do not address its use as a prophylaxis or in patients with less severe SARS-CoV-2 infection managed in the community.” So far as I know, the advocates of the HCQ et al treatment don’t advocate its use for people who are already in a really bad way, but instead very early on, before the virus has really taken hold, such as was done in the Henry Ford study.

On 2020-04-30 04:10:22, user Tom Turek wrote:

So.. how much Vit D3 is good? To get the optimum blood level of Vit.D3 of 60nano gms/mL of blood,we need either 5 minutes of sun a side, between 10am and 2pm in the lower latitudes. Longer time for darker skins.. or supplement with 8000IU of D3/d, BUT at t his high dose´ also Vit. K2 to stop calcium pumping into artery walls. )Ignore the outdated, idiotic RDA of 800iu of D3.. and avoid the synthetic, toxic D2 in nearly all multis.

On 2021-08-28 11:49:17, user Doug Truitt wrote:

So if one lives in Kentucky previous infection is less protective than vaccination - https://www.cdc.gov/mmwr/vo... - and if one lives in Israel previous infection is more protective than vaccination (this study). I'd be interested in discussion as to why these two studies are at odds with one another.

On 2021-10-15 21:36:23, user Mary V wrote:

Please add results for the unvaccinated individuals who recovered from Covid prior to Feb 28, 2021. How does their risk of symptomatic infection during the delta uptick compare with the individuals who were fully vaccinated by Feb 28, 2021 and didn't have a positive Covid test before 6/1/21? Did one shot of the Pfizer vaccine improve the immunity of that group during the delta uptick? Other studies have shown a very strong immunity for those who have recovered from Covid symptoms. The data in this paper only supports a vaccine recommendation for people who've had asymptomatic cases of Covid. Hence the data for those who've had symptomatic cases needs to be added or your conclusion about the vaccine recommendation should be qualified.

On 2021-08-26 23:02:47, user Klaskow wrote:

So when is the Peer review going to be scheduled?

On 2021-09-07 14:08:23, user Terry Davison wrote:

1979... what about recently.

On 2021-09-10 14:50:45, user disqus_ErZz7QdMBI wrote:

Its not about pretending, its that the Delta variant is literally 1000x more transmissible than the original strain. And as it accounts for over 80% of COVID cases in the US of course we have a lot more breakthrough cases. The new variants Delta and Mu are not the result of a "leaky vaccine" though. They are the result of the virus running rampant in countries where there was no vaccine.

On 2021-10-22 14:39:09, user 50% Leaving wrote:

Speaking of bias; yours seems to have blinded you to the multitude of methodological flaws.

1) that's not what their finding suggest at all<br /> 2) That's not how the vaccines work nor why there's increased observed virulence; the protein spike target of the mRNA vaccines was chosen explicitly because it's a target the viruses cannot readily mutate (as it's crucial in how the virus infects host cells)<br /> 3) That's not the case total nor death total (1.3 million cases reported, 8000 deaths reported)

Yes, it's interesting research; but peer review process is still underway, as the author's stated at the bottom of the article "preliminary scientific reports are not yet peer reviewed; and as such should not be considered conclusive"

On 2020-04-22 02:06:58, user Alexey Karetnikov wrote:

You got the virus taxonomy wrong: 229E and NL63 are members of the genus alpha-coronavirus.

On 2020-04-22 22:16:42, user Greta Bauer wrote:

Hi. You seem to have confused an incubation period range with a 95% confidence interval for mean incubation period. These are not the same thing.

On 2023-10-24 02:18:23, user CDSL JHSPH wrote:

Dear Dr. Bi et al,

Thanks to your work on influenza, which has provided a new proof that the residual repeat vaccination effect might be explained by different rates of subclinical infection between repeat and non-repeat vaccinees via two proposed mechanisms, the infection block hypothesis and enhanced vaccine immunogenicity and protection post-infection.

As a reader who doesn't know much about the field ,I can give you some reading feedback for your reference.<br /> First I think your article provides three important pieces of evidence. <br /> 1,Repeat vaccinees were vaccinated earlier in a season by one week.<br /> 2,Clinical infection influences individuals’ decision to vaccinate in the following season while protecting against clinical infection of the same (sub)type.<br /> 3,Adjusting for recent clinical infections did not strongly influence the estimated effect of prior-season vaccination.<br /> 4,Adjusting for subclinical infection could theoretically attenuate this effect.

On the basis of your good work, I would like to offer a bit of advice for readers who are not experts in this area. First is the article structure section. I hope this provides some perspective to help you publish. First of all the name of the title may be too long for non-specialized readers. It may lose some of the attention. Furthermore the explanations of within-season waning, recent clinical infection, and subclinical infection could have come in the INTRODUCTION instead of the METHOD before being mentioned. Another thing is that I think you could put in the conclusion that has some summarizing words underneath the fig so that readers might find them easier. Also the part you put in the appendix about theoretical modeling some of it is necessary for understanding the model, if you could summarize it necessary and put it in the body methodology would help understanding.

The next aspect is about research, first of all your work is very relevant and on this basis you are perfectly placed to capitalize on this aspect. First of all you can try to find some way (like sampling) to compare your theoretical model with the results of the data response you got, so that you can speculate about the effect of the vaccine in reality and the possible number of Subclinical infections.

After that comparing your model with real data would be an interesting aspect. And this can emphasize the correctness of your model and increase the credibility of your article. Although overall this article has been highly relevant with enough realistic data.

However, on this basis, one can consider whether the data from different regions (the five regions sampled) are very different? For example, are the probabilities of tuning into a vaccination strategy similar in different regions, and do repeat vaccinees in each region tend to get vaccinated earlier in the flu season than non-repeat vaccinees? These comparisons of data from different regions can be added to the article as they relate to the reproducibility and generalizability of your model's and conclusions.

Finally, thank you for this article, which provides very good evidence for the causes of Reduced effectiveness of repeat influenza vaccination, and this article has the advantage of incorporating a lot of details that were not considered in previous studies and provides a good interpretation of the errors, providing new ideas and theoretical models for the field. And I personally learned a lot of research ideas from you through this article, thank you for your work!

On 2020-12-15 10:43:49, user NK wrote:

Re: article pre-published at https://www.medrxiv.org/con...

There are several methodological problems in this study.

1. Findings that suggest increased ORs among primary school teachers, child care workers and secondary education teachers are not properly presented and discussed

The summary states: "Teachers had no or only moderately increased odds of COVID-19". This finding is mentioned several places in the text of the article. Teachers are repeatedly referred to as having a low risk, even when the results for teachers show a significant increase in admissions and borderline significant increase in infection rates. Quotes: «First, our findings give no reason to believe that teachers are at higher risk of infection», and in the conclusion: “Teachers had no increased risk to only a moderate increased risk of COVID-19”. We wonder why the authors find it important to repeatedly mention this the result for teachers when the results for the last period does not exclude a substantial increased risk for teachers, whereas occupational groups with lower risk than teachers are not mentioned in the summary.

The part of “Supplementary table 1” shown below does not provide a basis for such a conclusion that teachers are a low risk group.

The OR (95% CI) for 1) primary school teachers 2), child care workers and 3) secondary education teachers were 1.142 (0.99-1.32), 1.145 (1.02-1.29) and 1.095 (0.82-1.47) respectively. The upper confidence limits are does not exclude 29 % to 47 % increased ORs, which represent substantial increases.

Concerning the results on the risk of admission, it is stated: «None of the included occupations had any particularly increased risk of severe COVID-19, indicated by hospitalization, when compared with all infected in their working age (Figure 3, S-table 2), apart from dentists, who had 7 ( 2-18) times increased odds ratio, and pre-school teachers, child care workers and taxi, bus and tram drivers who had 1-2 times increased odds ratio”.

This finding is not discussed or mentioned in the summary, even if the findings were statistically significant for pre-school teachers as well as for child care workers.

1. The study periods include periods when the schools were closed and include no period with high infection rate among children and youths.

It is not to be expected that teachers have higher infection rates than the average working population in periods when school are closed and when the infection rates are low in the age groups 0 - 9 and 10 -19 years. This problem is not discussed in the paper. Schools were closed from 12 March to 27 April. For a majority of the schools, holiday started from Friday 19 June.

The first study period lasted from February 27 to July 17. Thus, schools were closed for over 70 days of the first study period of 139 days. The infection rates in children at school age in the first study period were rather low (3.6 per 100 000 children per week between in the age group 10 -19 in week 19, 1.1 per 100 0000 children per week in week 25). In the last study period, the infection rates varied between 7 to 17 per 100 000 per week in the age group 10 - 19. Even if these rates are much lower than later weeks that were not studied (after week 42), the results from this second part of the study suggest an increased risk for teachers.

Thus, the infection rates among children started to increase from week 43, after the end of the study period. By not including this period, the study design excludes the possibility to detect if these high rates among pupils could be related to increase infection rates among teachers.

It is a problem that the results from this pre-published study has been quoted in the media and referred to as if teachers have no excess risk, or even possibly a reduced risk at the time that several municipalities were to decide what type of restrictions at schools should be introduced to reduce the risk of transmission among school children, see https://www.barnehage.no/korona/ny-forskning-nei-barnehagelaerere-har-ikke-okt-risiko-for-smitte/211143

On 2021-08-04 08:41:37, user ingokeck wrote:

Dear authors,

Thanks for this fascinating study! I am really surprised that you only found 6+3=9 cases in your sample.

Your sample corresponds to 60/330=18% of the population of the US. If we assume that it is representative, this means in the whole US there were only ca. 50 cases of myocarditis in 12-17y old in total. Using the estimation of the CDC that 37% of all kids were infected (see https://www.cdc.gov/coronav... Table 2), this is a rate of just 50 in 30,000,000*37%, thus 4.5 cases per million infected.

So, the risk of myocarditis from Covid-19 is 4.5 cases per million infected and factor 100 smaller than what you report in your result, probably because your cases in the dataset are mostly hospitalized which make only a very small subgroup (ca. 1%) of all cases in this age group.

Compared to the risk from the mRNA vaccines of 10 per million (your source) or rather 333 per million based on the results from Israel (1:3000), the risk from the vaccination is 2 to 100 times greater than the risk from a Covid-19 infection.

I would appreciate it very much if you could correct your calculation to the correct base infection rate as given by the CDC, and also discuss why you assume a smaller vaccine risk than the data from Israel suggests.

On 2021-08-21 19:03:01, user Jonathan C wrote:

Hello,

Thanks for an interesting analysis. CDC estimates a far higher infection rate (36.77/100k, <br /> https://www.cdc.gov/coronav... "https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burden.html)"), <br /> at a similar rate for the 0-17 y group, although they do not seem to show data for the 12-17 y group).

Am I correct in interpreting your assumption that the infection rate for the <br /> investigated COVID-19-related period was at a far lower <10%? (and that 2.5% of all COVID-19 cases should represent males aged 12-17)

Or is there some information missing regarding your analysis?

On 2021-09-11 19:31:00, user Rikk wrote:

CDC has done a good job in pinpointing covid risk factors. Age and high BMI stood out. Other studies have confirmed vitamin D deficiency to increase severity of disease. We can obviously ignore age in a study of children. But why is correlation to BMI not included? It is easy to obtain. Vitamin D status should be included where available.<br /> Such a high level view becomes crude as the individual variation of risk factors has a major impact typically. Only with refinement of data can good conclusion be made, I think the work should strive to use the CDC defined risk factors as much as possible and as an overlay to analyse the risk of myocarditis for each CDC defined co-morbidity. Especially if the study has an intent to be a guide for any sort of intervention.

On 2020-07-09 21:41:42, user kpfleger wrote:

Thank you for this study! Suggestions to improve the manuscript:<br /> (1) In the PDF version linked here from medRxiv (as of 2020/7/9), p.1 states the multivariate infection OR as 1.45 but p.5 & table 3 list it is 1.50. Minor discrepancy but good to get the p.1 results summary correct.<br /> (2) You imply in the conclusion that most of the 25OHD test results were recent, such as upon presentation to health services for illness, but it would be helpful for you to characterize the dates of the 25OHD tests in your cohort. (Eg, so we know they aren't as old as the 10+ year old UK Biobank results.)<br /> (3) It would be helpful to have the descriptive statistics for demographic and clinical characteristics for the hospitalized vs. non-hospitalized COVID-19-P patients---analogs of tables 1 & 2 stratified by hospitalized or not. I'm not even sure you say how many were hospitalized.<br /> (4) You gave the multivariate adjusted OR for infection and the OR for hospitalization given infection. It would be nice to state the adjusted OR for absolute risk of hospitalization (not specifically given infection) as this is perhaps more meaningful for public policy.

I look forward to a couple more analyses like this from other large HMOs or government insurers around the world. I also look forward to data on post-hospitalization measures of severity (eg, ICU/ITU admission, fatality) in population cohorts this large.

On 2021-02-26 16:48:14, user Jason Beugnet wrote:

what journal is this published within? Just trying to make a citation of this

On 2021-03-26 18:00:33, user ayman alqunneh wrote:

Dear Mr. external reviewer without name, <br /> I think marketing the campaign through the main researcher's social media accounts is an advantage, not a disadvantage. This vast number of participants in the survey gives it the volume to be a large-scale multinational survey. And this is not new, as this is the way how we as professors encourage people to participate in surveys and questionnaires. So your claim is not valid unless you prove that the researchers guided and misled the participants while filling the survey. On the other hand, a large number of participants in this survey makes it impossible to be biased. I hope next time to be ethical in your criticism and criticize the research itself, not the researchers!

On 2021-03-13 07:53:06, user Motaz wrote:

Thank you for such a great work. I just want to point out in the list of figures at the end, "Fugure 4" is misspelled.

On 2021-10-19 17:10:24, user Jeremy Gustafson wrote:

It's too bad they didn't break out a 5th group of previously infected and not vaccinated to compare their immunity similar to the study that was done in Israel.

On 2020-12-24 10:34:39, user Aroop Mohanty wrote:

The review is one of the first of its kind at a national and international level. It is good to see that mobile health intervention found effective in improving maternal health outcomes. The review and meta-analysis did very well and need of hours for developing country including India. the methodology and search strategy used very clear and crisp and elaborated in detail. I am impressed that the use of Mhealth intervention improved maternal outcomes. <br /> The use of the PICO approach and exclusion of research done in developed countries have a direct implication of the work in low and middle-income countries like India.<br /> The research question is clear and concise<br /> I am highly recommending such kind of work to improve maternal and child health indicators in developing countries.

On 2021-10-26 09:42:04, user Stephen Hinkle wrote:

I think this study calls for an important discussion about how we approach COVID-19 in the future. I think that it is clear that people can get this more than once. Other studies have shown that vaccine immunity is not lifelong either. I think we need to INFORM THE GENERAL PUBLIC OF THIS TRUTH and have a public policy discussion where the public is invited to participate on how the public wants to confront COVID-19 longer term going forward. It is likely that this will be an endemic virus (this is the conclusion of many top public health universities including Brown, Harvard, Stanford, Johns Hopkins, University of Minnesota, Imperial College London, University of Alabama at Birmingham, University of Arizona, University of Sydney, University of Queensland, Oxford, and others). Many countries have abandoned their "Zero Covid" strategies as well realizing this including Australia, New Zealand, Vietnam, Thailand, Singapore, and others. This study covering Iran shows that people got infected many times.

Do we stay in lock down and abandon some activities and pleasures in life forever possibly leaving businesses permanently closed or forcing everyone to say their last goodbye to our friends, abandon all group activities, sports, performing arts, dating, and our pleasures in life forever in an attempt to stay alive or stop the virus? Do we open up and accept the risk of ongoing community spread of COVID-19 and keep getting booster shots for individual immunity and new variants? Should getting vaccines be mandatory or an individual level decision? How do we protect the immunocompromised and those who are more vulnerable or who the vaccines do not work well on? Do we do a massive COVID-19 testing operation and try to eliminate the virus through daily tests and quarantine people if they are infected an allow the others to go on with normal life activities? What level of death and disability should society choose to accept to have the levels of freedom of movement and/or non-household member social interaction we want in the future if the COVID-19 virus will be endemic? Should shuttered sectors of the economy be allowed to reopen or not? Should in-person schooling continue or not? Should masks be required indefinitely or should it be optional or not required?

I think it is time to start a policy conversation with the GENERAL PUBLIC to determine what they want the un-perfect pandemic endgame to be in terms of living with the virus and going on with life as safe as we can but it is likely the day-to-day risk will not be zero. It is clear to me based on all the recent evidence from this study and all the current data trending in other recent studies is showing that COVID-19 will become ENDEMIC and that this pandemic is going to have a social ending as opposed to a eradication or herd immunity outcome most likely. But the real question now is what will a divided public tolerate in terms of COVID-19 policy longer term and what is the public health end goal now? Maybe it is time to ASK THE GENERAL PUBLIC FOR IDEAS here.

Personally, I think that the COVID-19 pandemic is another case of humans showing a poor record of eradicating diseases.

On 2021-11-16 17:54:49, user C D wrote:

Why do we keep thinking herd immunity can be achieved for every strain? Isn't it normal to have new strains that people aren't immune to yearly? Iran's covid cases have currently plummeted, what happens if they stay there? Does that mean herd immunity has been achieved?

On 2020-04-24 05:24:07, user traxw wrote:

This is an older thread. NYT article today about blood clotting being a factor in prognosis that had not been previously considered or known about. It would be interesting if there's a correlation between type and susceptibility to clotting as a result of infection.

On 2021-12-27 04:46:23, user Wes Higbee wrote:

So, if I get vaccinated, I markedly increase my chances of infection?

On 2021-08-18 15:39:23, user Lincoln Sheets wrote:

This analysis seems to present a paradoxical finding, that there is an inverse relationship between early deaths from cancer and early deaths from CVD. If high animal-source diets lead to more early deaths from cancer, does that make it impossible for those same persons to die early from CVD? This is a detailed big-data study that deserves reivew, publication, and expansion by additional researchers.

On 2024-10-13 17:50:05, user Tom Hagan wrote:

Consider that niclosamide is highly lipophilic (LogP of 3.91) and the distal rectum’s inferior and middle veins have been employed to systemically transport many rectally administered, insoluble drugs. Would niclosamide have greater bioavailability if simply compounded as rectal suppositories? Apparently - from searching the journals- this has never been attempted.

On 2020-05-20 06:50:14, user Chris Valle-Riestra wrote:

Thank you, I can see that this is a very important finding for understanding the development of the epidemic in any nation, region, or city. That heterogeneity in susceptibility would have this effect can be understood intuitively, as soon as one really starts to think about it. Determining an average R nought for an entire nation, and making projections based on that alone, plainly doesn't tell the whole story.

A simple thought experiment will demonstrate this. If an entire population is split into two sub-populations of equal size, and the individuals in one of the sub-populations all have low susceptibility, effective R just for that sub-population can be well below 1.0, in spite of a generally high virulence of the virus. Very few in this sub-population will ever become infected. The other half of the full population will be highly susceptible, and a substantial majority of that sub-population would be expected to become infected over time. Adding it all up, something well under 50% of the total population will ultimately become infected, and herd immunity will have been achieved.

Recent small serological studies around the U.S. have typically indicated a middle-of-the-road level of infection, ranging between perhaps 6 and 30 percent from place to place, many weeks into the epidemic. This has struck me as perplexing. Based on the usual naive model of the development of an epidemic, one would have thought it likely to find either (1) a very low level of infection, such as under 5 percent, implying great success in suppression efforts, or (2) infection levels moving steadily past 50 percent, implying a high R nought that suppression efforts were inadequate to suppress. Basically, either suppression would work or it wouldn't. It would be surprising to find that that the virus had enough power to infect a major fraction of the population, carrying a big head of steam going forward, and yet be able to be halted that late in the game.

Your finding points to a likely explanation for this phenomenon. It suggests to me a likelihood that the epidemic in the U.S. has been working its way through the most susceptible sub-populations, not successfully checked, but that it has made little progress in infecting less susceptible sub-populations.

I think it should be recognized that to the degree that an individual's susceptibility is based on his social conditions, that may change over time. An individual living far out in the country may have little connectivity, and therefore little susceptibility. If he moves into the heart of a city, that may change. This implies that herd immunity is likely to "erode" over time. COVID-19 is likely to remain endemic and to continue to cause a low level of disease, serious and otherwise, for a long time to come.

Be that as it may, there's a strong likelihood that public health officials and political leaders have been seriously misinterpreting the progress of epidemic. This has major implications for public policy choices. Further research is urgently needed, and decision makers need to develop a more nuanced understanding. They are currently making weighty decisions based upon a probably badly flawed model.

On 2020-01-31 21:48:51, user Carl Asplund wrote:

Some things I'm wondering about: <br /> Line 61 - "smaller" should be "larger"<br /> Line 72 - The first date (year) is wrong<br /> Lines 87-88 - What are the additional modelling assumptions made here? The model on line 45 doesn't support R values less than 1.

On 2022-01-19 16:21:44, user xavier wrote:

Hi, fig 4/inset d, shouldn't line one of the first line read "infection WA1" instead of "infection B1"? <br /> Thanks

On 2021-08-06 21:57:16, user dustinst22 wrote:

That's due to vaccination. Virulence is measured against those without seroprevalence.

On 2023-11-26 10:25:15, user Chris Dye wrote:

Just for fun, you might also be interested in the long-forgotten... Epidemiology Infect. (1995), 115, 603-621 Measles vaccination policy

On 2020-06-17 15:34:18, user Dr. Amy wrote:

NSAID users were MUCH older and more likely to have comorbidities associated with adverse outcomes. Can't make too much of this without controlling for those big confounders.

On 2020-08-29 19:38:23, user Davers wrote:

So much missing from this study. Why weren't blood levels of en vivo vitamin D, magnesium, etc taken and reported per patient before, during, and after DMB? This is helpful to know when and where this protocol will be most beneficial. Significant shorter hospital stay was identified as an outcome, but no numbers are given, nor was given the criteria by which patients were released. Why does the table appear to be incomplete with regard to those needing ICU care (see LB comment). Studies like this could be transformative but little mistakes like these (even if they are just reporting errors) make them too easily dismissed as being low quality.

On 2021-02-01 11:20:07, user Fjortoft9 wrote:

Given that the study is assuming the rate of vaccinations will be around 1m a week in January, rising to 2m by February I’m afraid it doesn’t seem to be very useful. <br /> We know now that the actual rate of vaccinations in January was more like double that and the rate in the last week is well over 2.5m. That difference would completely change the modelling and it’s disappointing that you didn’t model the impact of a faster vaccination rollout.

On 2020-07-20 23:40:04, user Patrick wrote:

tom jarman said that they recommend the use of masks. What he was saying is that the researchers are not clear on why they still recommend masks

On 2021-05-28 19:46:09, user Stop the Insanity wrote:

Russ - these studies (yes, I read them) are mostly estimates of projected benefits from mask wearing based on various assumptions. We assume wearing masks stops 20% of the potential virus transmission (based on some mask transmission analysis or other data) and then the study projects the benefits of wearing masks, based on that assumption. These were fine when they were performed, mid-to-late 2020, but the analysis being released now are based on actual observed virus transmission rates and control group studies of actual people and virus transmission.

While I am not faulting the earlier studies, I also am not going to continue to see them as conclusive given what we are learning. If an earlier study projected that mask wearing would reduce the spread of the disease by 20% or more, yet actual analysis of mask wearing indicates that there is no difference in infection rates between those that wear masks versus those that don't, then it makes little sense for me to continue to site the earlier study since its projections didn't hold up to reality.

But hey, maybe that's just me.

On 2021-05-28 12:42:16, user Barbara Elizabeth wrote:

For those who have commented here that wearing the mask (mandate or not) is what lowers numbers... another study on NIH site disagrees (and you've surely heard of this) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7680614/. The conclusion states: <br /> "The existing scientific evidences challenge the safety and efficacy of wearing facemask as preventive intervention for COVID-19. The data suggest that both medical and non-medical facemasks are ineffective to block human-to-human transmission of viral and infectious disease such SARS-CoV-2 and COVID-19, supporting against the usage of facemasks. Wearing facemasks has been demonstrated to have substantial adverse physiological and psychological effects. These include hypoxia, hypercapnia, shortness of breath, increased acidity and toxicity, activation of fear and stress response, rise in stress hormones, immunosuppression, fatigue, headaches, decline in cognitive performance, predisposition for viral and infectious illnesses, chronic stress, anxiety and depression. Long-term consequences of wearing facemask can cause health deterioration, developing and progression of chronic diseases and premature death. Governments, policy makers and health organizations should utilize prosper and scientific evidence-based approach with respect to wearing facemasks, when the latter is considered as preventive intervention for public health."

On 2021-08-22 16:27:15, user CJS wrote:

Then why is Fauci saying a mask doesn’t protect the wearer? Your assumption that the virus must attach itself to a large respiratory droplet is naive.

On 2021-05-30 05:39:10, user FatBoy “JD” Diesel wrote:

Seeing people going back and forth here, but I'll just point this out:

If wearing masks are really that effective, as some claim, how are medical staff getting infected on the job despite the extra medical gear, protection, and protocols? The overwhelming majority of you can only provide conjecture at best. If medical staff adhering to strict sanitation protocols still get infected by a respiratory illness on site, what makes anyone believe that less effective equipment, or less physical coverage provided by said equipment, would be effective for a general population less equipped to adhere to stricter sanitation protocols?

On 2021-05-29 05:03:42, user Mike Stevens wrote:

Well, your argument would just verify my point...it’s the wearing of masks that cuts Covid transmission, not whether there is some advice in place to do so.<br /> But of course one needs to wear them properly, or they may not work.

An suitable analogy might be as follows:<br /> “Installation of clean water and sewage facilities does not cut dysentery risk”<br /> If one finds that a significant number of the population are refusing to practice basic hygiene and are still defecating in the wells used for drinking, do you think the study conclusion should be “Public health interventions do not stop the spread of dysentery”, and for governments to cease to provide tap water and toilets for their citizens?

PS: The reuse of surgical masks depends upon their contamination. They are used for 4-6 hours continuously in hospital settings. It’s advised that you wear a new mask if as a member of the public you are taking one off and then need to remask. There is an elements of common sense here... if I am going into say Tesco’s for a few minutes, then into Boots 5 minutes later, I might for the sake of convenience just put on the mask again, assuming I have kept it clean in the interim. I notice some people “wearing their mask” in open air spaces in towns, usually because they are going from one shop to the next a few doors down. Taking the mask off between those times seems more trouble than it’s worth, so they leave it on. This week I was dropping in at the petrol station attached to my local Tesco superstore, so I didn’t bother to take off my mask, drove 300 yds in the car while wearing my mask, then got petrol. No big deal. I didn’t die from hypoxia, but no doubt some antimasker laughed when they saw me wearing the mask in the car.<br /> Just use your common sense, I say. There’s not much of that about these days.

On 2022-06-24 22:44:55, user S. W. wrote:

This preprint has been published in Emerging Infectious <br /> Diseases at https://doi.org/10.3201/eid2808.220617.

On 2020-11-03 14:18:18, user Jonas Svensson wrote:

Paper now published in Journal of Thrombosis and Thrombolysis:

https://doi.org/10.1007/s11...

On 2021-06-10 09:33:28, user Marcelo Mazocco wrote:

Stephen, can you assure that not even around 7 out of the 224 patients to whom were given HCQ did get included in the "success" group by cumulative dosages (the 18 survivors that took >3g HCQ + >1g AZM) JUST BECAUSE they survived and therefore had more time to consolidate those dosages, and not the other way around?

Because if that is the case, I believe you got a problem of statistical significance. That would leave us with just 3 survivors above general average. And by 7 out of 224, that might be the case.

Another issue: 87% of all patients received HCQ in some form but what was the percentage of those in the survivors group? I couldn't find the number but I suspect this result is something to look at. It should be higher than 87% if any form of it works in any way. As is also something to look at the randomized double blinded studies that compare those who took HCQ+AZM to those who didn't, in a more broad way, so that we can compare the effects. I suspect that this study not being blinded or double blinded could have also skewed the results by some amount, and that is another issue.

One more issue: I find it quite weird that you had 78% death rate when the average north american death rate for intubated patients of covid is around 46% for what I've seen. Even using a few medications that supposedly work. There must be something else at play here, probably also skewing the results.

On 2021-06-10 14:55:18, user J.A. wrote:

The statistical analysis is invalid suffering from an very clear immortal time bias. This is really basic stats 101. There is a reason this pre-print has not been accepted in a peer-reviewed journal.

For a Kaplan-Meier survival analysis, groups need to be defined at baseline. For persons to have received >80 mg/kg cumulative HCQ dosing (where a standard 600mg dose could is ~8mg/kg/day), persons need to survive for 10 days. So Figure 2, compares those who survive to receive 10 days of HCQ versus all others This is an immortal time bias.

As 88% of people overall received HCQ, likely a large proportion of those 40% who died before day 10 received HCQ (at max only ~12% could not have received HCQ).

The authors should re-do their analysis using a Cox-regression with a time-dependent covariate for receiving HCQ.

On 2024-06-21 21:19:36, user WILLY CESAR RAMOS MUÑOZ wrote:

Published in BMC Cancer: https://bmccancer.biomedcen...

On 2023-06-07 17:10:10, user Nathan Pearson wrote:

Also: in adding such needed apples-to-apples comparison of bivalent vs. monovalent peer boostees (i.e., with 3+ doses in each otherwise background-similar group), can the authors take care to control for time since last dose?

I.e., if bivalent boostees got boosted at most k months before data freeze (so were tracked for severe COVID for <=k months after last dose), likewise tally cases of severe COVID in monovalent boostees <=k months (but not longer) after their last dose.

Thanks.

On 2020-04-04 19:45:40, user Ibraheem Alghamdi wrote:

That is the thing with ecological studies, they are good at generating hypotheses and interesting, but, they prove nothing.

On 2021-12-25 11:00:44, user Jeffrey_S_Morris wrote:

These results are strange indeed.

The fact that this time period is so short, and that delta was still so dominant during this period, makes these results very difficult to interpret given the competition between delta and omicron

Given the already well documented immune escape properties of omicron vs vaccination, previous infection and monoclonal antibodies, yet the strong transmission advantages of delta, it is possible (likely?) that the infections from those with immune protection are predominantly omicron, but for those without immune protection it is predominantly delta. This effect could strongly reduce omicron specific infection rates in unvaccinated and have a strong effect on VE estimates.

Too bad they removed previously infected from these data and didn’t consider them a stratified cohort. If this competition effect was a major factor here, then one would also expect the cases among previously infections would also be dominated by omicron, and previous infection would also be deemed to have “negative effectiveness” vs omicron. These data would give us an idea of whether the negative effect is a component of vaccination or an artifact of this competition effect.

If omicron eventually dominates and delta disappears, then this competition effect would disappear and we’d have a clearer sense of VE of vaccines vs omicron outside of this competition effect.

But there is also a chance that delta has an inherent transmission advantage that will cause it to remain dominant in unvaccinated while omicron’s immune escape makes it dominate among vaccinated and previously infected.

In that case we might see delta and omicron coexist in some sense.

We will have to watch these data unfold in the coming weeks and months

On 2021-12-25 21:50:03, user Daniel Song wrote:

This is just a poor quality paper

On 2020-06-29 02:58:37, user David F. Priest wrote:

Study has not been peer reviewed and was funded by Suez which has a joint venture in China with the state-controlled China Everbright International Limited.

On 2020-04-19 12:20:06, user Donna Marie Seyfried wrote:

Would be good if he used spell check. I’ll wait with my comments ntil it’s peer reviewed .

On 2020-05-27 02:53:57, user Divalent wrote:

Are case data the date that test results were reported to the public, or the date the lab determined the test result, or the date of test sample was taken, or the date of first symptoms? (I'm trying to get a handle on what sewage detection tells us, and how it can be used. I.E., how much of the 7 day offset is due to asympt shedding, vs test-processing delays vs test result reporting delay, vs time from sympts to time of test.)

On 2021-03-18 06:40:29, user CD wrote:

I have not read the full paper. Cautious comment: Recruitment between July and December is too large an interval. For example, if in one region most of the recruitment was done in July and in another most aas in December, this will affect the results.

On 2021-07-16 09:03:00, user Ashish Agrawal wrote:

I am myself fully vaccinated with Covaxin with no side effects and have an IGG antibody score of 150.00 which is apparently enough as per many studies. <br /> While vaccinating with Inactivated virus vaccines one should maintain proper isolation from those who're not vaccinated with it. I don't think travelling between the dose 1 and dose 2+ 14 days was a good idea. Vaccines work but they take time in changing environment

On 2020-03-20 17:02:20, user Kevin Hamill wrote:

An editing suggestion:<br /> This manuscript will be read by new media/journalists therefore I would encourage more careful use of the term "significant".

For example where it says "blood group A had a significantly higher risk for COVID-19 " if that was quoted then people would hear this as "blood group A had a much higher risk for COVID-19."

In the same sentence, I would write:<br /> blood group A had approximately 20% higher risk for COVID-19 (odds ratio-OR, 1.20; 95% confidence interval-CI 1.02~1.43, P = 0.02). <br /> [and equivalent changes with the other phrases].

Note that as you have already stated the p values, using the word significant has no added value, it only provides a source for ambiguity.

On 2020-03-23 16:59:18, user Rintaro N. wrote:

?“The distribution of ABO blood groups in blood donors... (ref.7)” might not be a good control. It can be different from “true” distribution.?

?cf.?<br /> ?Blood Type and Blood Donation Behaviors: An Empirical Test of Pure Altruism Theory<br /> https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3171957

On 2025-06-07 04:13:42, user Takenobu Shimada wrote:

https://doi.org/10.1016/j.ijcard.2025.133464

On 2020-06-03 10:57:43, user Sebastian Fiebiger // medizin+ wrote:

Interesting study that provides us with further indication of the transmission pathways of SARS-CoV-2 and their significance in the context of the pandemic.

There is currently no "Aha!" kind of study. But there are many small "building blocks" that provide an overall picture.

Thanks a lot to the team!

Warm regards from sunny Berlin,

Sebastian<br /> medizin.plus

On 2020-04-18 02:01:37, user mendel wrote:

First, he picked the county that had the earliest cases in California and had the outbreak the first, ensuring that the population would be undertested. This means that it's likely that every other county in California has fewer unregistered infections than Santa Clara.

Second, study participants were people who responded to a facebook ad. This is a self-selected sample, and this property completely kills the usefulness of the study all by itself. This is a beginner's error! People who think they had Covid-19 and didn't get tested or know someone who did are much more likely to respond to such an ad than people who did not. (By comparison, the Gangelt study contacted 600 carefully chosen households per mail, and 400 responded. Still somewhat self-selected, but not as badly.)

Third, age is the one most common predictor of mortality. He did not weigh the results by age, and old people are underrepresented in the study. Anything he says about mortality is completely useless if we don't know how prevalent the infection was in the older population. (In Germany, cases show that the prevalence among tested older people was low initially and took a few weeks to rise.)

Fourth, instead he weighs prevalence by zip code--why? This exacerbates statistical variations, since there were only 50 positive results, and Santa Clara has ~60 zip codes. If you have a positive result fall on a populous zip code by chance where only a few participants participated, then the numbers are skewed up. They must have seen this happen because their estimated prevalence is almost twice as high as the raw prevalence.

Fifth, the specificity of the test is "99.5% (95 CI 98.3-99.9%)". This means that theoretically, if the specificity was 98.5%, all of the 50 positive results could be false positives, and nobody in the sample would have had any Covid-19. This means the result is not statistically significant even if the sample had been well chosen (which it wasn't). (It's not even significant at the 90% level.)

Sixth, they used a notoriously inaccurate "lateral flow assay" instead of an ELISA test and did not validate their positive samples (only 50) with a more sensitive test -- why not?

Seventh, The Covid-19-antibody test can create false positives if it cross-reacts with other human coronavirus antibodies, i.e. if you test the samples of people who had a cold, your speficity will suffer. Therefore, a manufacturer could a) test blood donor samples, they not allowed to give blood if they have been sick shortly before; b) test samples taken in the summer when people are less likely to have colds than in March.

To state the previous three points this in another way, a large number of positive results (a third if the specificy is actually 99.5%, but probably more than that) are fake, and depending on which zip codes they randomly fall in, they could considerably skew the results.

On 2020-04-23 02:02:17, user Joseph Cole wrote:

A couple of weeks ago I drew a scatterplot of confirmed cases (per million population) versus tests made (also per million population) using country-level data and found a pretty neat linear relationship (in log scales). Just extrapolating from the fitted model I estimated that 3.1% of the population would be found infected if everyone was tested. I know it's a wild extrapolation, but do you think it works as a "quick and dirty" method to obtain a reasonable ballpark figure?<br /> https://uploads.disquscdn.c...

On 2020-04-20 18:38:21, user svenj wrote:

Academically interesting and a nice statistical exercise, but consider that COVID 19 is now the second leading cause of death in the US (https://www.dallasnews.com/... "https://www.dallasnews.com/opinion/editorials/2020/04/19/the-number-are-in-covid-19-is-worse-than-the-flu-and-is-now-a-leading-cause-of-death/)") and this is with social protocols in place. Kind of immaterial how many people statistically might have, or have had it. It's killing people at a faster rate than almost anything else right now. I can only imagine if you back off on distancing, those who don't know they have it will pass it on to more for whom it is a problem. Doesn't sound like a good idea.

On 2020-04-19 08:32:24, user Matthew Markert wrote:

What is the specific data on cross-reactivity of other CoV strains on this Ab test (Premier), and what is the expected or known prevalence rate of those strains in the background population?

If that is unknown or unknowable, can you instead run PCR on all the tested Ab samples for other common CoVs? If you can rule out that as an underlying confounder, or can show that they are present in people who tested negative for SARS-CoV-2, it would strengthen the data.

As written, and also for other reasons stated elsewhere (including your reported false positive rates and potential to explain a section of the 50 cases), the true positivity rate remains unclear. https://uploads.disquscdn.c...

On 2020-04-18 07:25:12, user LCMB wrote:

This is a false premise. Almost every article written prior to the event had people begging to be included in the comments section. No one I know would turn it down. Having been exposed previously would give some comfort, and also allow one to venture back to work without the fear of getting it again (at least any time soon).

On 2020-04-22 12:09:29, user Three-legged Crow wrote:

Where’s zinc mentioned in this study?

On 2020-04-22 13:13:49, user NG wrote:

In this study some of the baseline characteristics of the patients - Pulse oximetry, Systolic Blood pressure, creatinine, Lymphocytes was worse for HCQ / HCQ+Azithromycin group *to *begin with as compared to No HCQ group. Which means the HCQ / HCQ+Azithromycin group patients were sicker than No HCQ initially. Thus cant conclude anything from the study IMO .

On 2021-12-28 04:07:37, user abworkshop wrote:

Epidemic is simultaneously 5% with symptoms or more of the population. We hardly have 1%.

On 2021-01-03 12:28:23, user Richard Weller wrote:

Very pleased to see this randomised double-blinded controlled trial of vitamin D on COVID outcomes, and not altogether surprised that it shows no benefit. Rule 101 of epidemiology is that 'correlation does not equal causation' which is frequently overlooked by those advocating Vitamin D supplementation as a universal panacea. The only way of convincingly showing causality is by interventions studies such as yours. Numerous meta-analyses now published in the major journals showing no benefit of Vitamin D on cardiovascular disease despite the strong inverse correlation between measured vitamin D and CVS outcomes. This needs to act as a cautionary tale for those determined to imply that the link between low vitamin D levels and poor COVID outcomes is necessarily causal. 'Vitamin D tunnel vision' also has the unintended consequence of stopping researchers looking for alternative mechanisms. We have shown that non-vitamin D forming UVA inversely correlates with COVID mortality in USA/England/Italy https://www.medrxiv.org/con... This suggests that there are mechanisms independent of vitamin D by which sunlight might improve outcomes, and that we should be looking for them.

On 2020-02-28 01:21:33, user RQ wrote:

It was an easy method to calculate the true T value and CFR without any indigestible mathematical formulas or models requiring severe calculating conditions. Actually, when different T was assumed, if it was smaller (bigger) than the true T, calculated daily CFRs would gradually increase (decrease) to infinitely near the true CFR with time went on. Left of true T is decreasing, right is increasing,so T could be easily determined, then the true CFR could be calculated. The calculated true CFR had accurately predicted the death numbers more than two weeks continuously

On 2021-09-10 01:38:54, user Tanner wrote:

A limitation to consider: A control and experimental cohort of "unvaccinated" and "Vaccinated" does not take into account a large population of previously infected individuals. This would likely have a large impact on the infection rate of both the vaccinated and unvaccinated cohorts and help guide current policies being passed.

On 2020-09-07 11:18:03, user Maria Elena Flacco wrote:

There are at least two major issues in the present pre-print manuscript:<br /> 1. <br /> The authors report that only one meta-analysis, published on MedRxiv, is currently available on the association between ACEi/ARBs use and severe/lethal COVID-19. However, at least two meta-analyses have been published on the topic in Medline-indexed journals (see for example Flacco et al Heart 2020 Jul 1:heartjnl-2020-317336 - doi: 10.1136/heartjnl-2020-317336);<br /> 2. <br /> All the included data seemingly came from observational, retrospective studies. The authors meta-analyze them computing RRs (and 95% CIs) using raw data, as if the studies were randomized controlled trials. This approach is not correct, as it is based upon unadjusted (and non-randomized) estimates: to account for the observational nature of the included data, a generic inverse-variance approach should have been used.

On 2020-05-06 20:25:49, user Frank Conijn wrote:

A furthermore well-written paper, in which in particular the section about the used dosage in vitro and in vivo is interesting.

But two crucial questions are not answered:

1. What determined that the HCQ group got it and the other group not? It's a single-hospital study, so that shouldn't be difficult to answer.

2. What other drugs, if any, did the groups get? HCQ has a strong antiviral effect in vitro, but in vivo seemingly also an immunomodulatory one, since it's effective in rheumatoid arthritis. So, several interactions may be possible.

On 2024-07-26 14:15:20, user Gauthier wrote:

This preprint has now been published in the Journal of Dentistry: https://doi.org/10.1016/j.jdent.2024.105130

On 2021-06-06 09:28:02, user Ulltand wrote:

4 days after one dose. What does it say? We know from other studies that 21 days after one dose protection is about 90 %. 14 days is a to short period.

On 2025-02-15 00:44:43, user HLA.Fan wrote:

The authors should not make up nomenclatures for discussing HLA alleles. "HLA-DQB1*57" is not a valid way of describing variation at amino acid position 57 of the DQB1 gene. Similarly, "DRB1*13", confuses the appropriate 1-field descriptor for all DRB1*13 alleles with variation at amino acid position 13 of the DRB1 gene. No where in this preprint do the actual amino-acid variants seem to be discussed.

Further, the authors refer to "two- and four-digit HLA alleles". The digit-based HLA nomenclature has not been in use for almost 15 years (see: DOI: 10.1111/j.1399-0039.2010.01466.x). The authors should familiarize themselves with and use, modern, field-based HLA nomenclature.

Finally, when the authors do use HLA allele names, the names must be used consistently. When one is referring to an allelic HLA gene variant, the entire allele name is italicized. When one is referring to a specific HLA protein variant, the allele name is not italicized. There is no situation in which, e.g., "HLA-DQB1" should be italicized when "*02:01" is not italicized.

On 2020-06-26 16:22:15, user disqus_XufFG9Zovr wrote:

Has this been adjusted for time?

Do the masks just slow the spread and delay herd immunity?

Is the total death in the community less over all time for mask wearers or is it just a technique to flatten the curve?

Mortality per day is not an adequate goal. Total death must be considered as well.

On 2020-05-22 14:10:19, user C'est la même wrote:

The very high female proportion and narrow recruitment strategy suggest these results might not represent the community as a whole, due to participation and response biases.

On 2020-09-22 03:41:51, user Direk Limmathurotsakul wrote:

This article has been accepted for publication at the Emerging Infectious Diseases (EID). https://doi.org/10.3201/eid...

On 2021-10-02 06:16:24, user Not Ready to Panic Dog wrote:

Since low Vitamin D levels are associated with increased incidence of cancer, heart disease, diabetes, and various auto-immune, neurological and inflammatory disorders, how did you account for the patients’ comorbidity influence on disease progression? https://pubmed.ncbi.nlm.nih...

On 2020-06-23 13:30:40, user Ralph Hawkins wrote:

Analysis of the RECOVERY trial pre-print data, looking only at non-ventilated patients together, not stratified by oxygen use. There is NO DEMONSTRABLE TREATMENT BENEFIT.<br /> Dex treated 360/1780 (20.2%) vs standard care 787/3836 (21.6%) p=0.2427

On 2020-04-04 14:12:44, user Ming Ming wrote:

This paper accepted by Lancet EClinicalMedicine

https://www.thelancet.com/j...

On 2020-08-03 13:49:53, user Michael wrote:

You may be interested in our research at the optimized boarding of passenger groups in times of COVID-19. We find that the consideration of groups in a pandemic scenario will significantly contribute to a faster boarding (reduction of time by about 60%) and less transmission risk (reduced by 85%), which reaches the level of boarding times in pre-pandemic scenarios.

The preprint publication “Analytical approach to solve the problem of aircraft passenger boarding during the coronavirus pandemic” is available here:<br /> https://www.researchgate.ne....

Additional information about common passenger boarding in times of COVID-19 are available here:<br /> Evaluation of Aircraft Boarding Scenarios Considering Reduced Transmissions Risks (https://www.mdpi.com/2071-1... "https://www.mdpi.com/2071-1050/12/13/5329)").

On 2021-01-01 12:19:30, user Ed wrote:

This article is a preprint and has not been peer-reviewed. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice.

..and yet it is being used for exactly that, ref. Department of Health and Social Care Joint Committee on Vaccination and Immunisation: advice on priority groups for COVID-19 vaccination, 30 December 2020 6th bullet-point of the introduction.

On 2020-06-06 13:05:56, user OxImmuno Literature Initiative wrote:

https://www.immunology.ox.a...

On 2022-08-04 17:25:32, user Paul Hunter wrote:

Did you include date or week number in your model? During the study period there was a dramatic shift in the proportion of tests positive in Portugal from about 1 in 4.5 to 1 in 2 and that could explain your findings of a 3 x greater risk of hospitalisation associated with BA.5 infection irrespective of the actual risk . If you did not include week number then I think your conclusions are probably flawed.

On 2025-03-26 17:07:23, user Paul wrote:

This article has been published in BMJ Global Health <br /> https://doi.org/10.1136/bmjgh-2024-016607

On 2020-05-03 14:25:00, user Geoff Turner wrote:

Comparing diagnostic tests like this is a classic signal detection problem. What is your d'? What's the d' for the nasopharyngeal test? What's the bias of each? This is the only way to know which test is most sensitive AND simultaneously least biased.

On 2020-09-13 07:55:39, user OxImmuno Literature Initiative wrote:

https://www.immunology.ox.a...

On 2020-12-21 00:56:39, user RP Rannan-Eliya wrote:

These systematic review findings are largely consistent with the findings from our global ecological analysis of 172 territories just released in Health Affairs that controls for multiple interventions and factors during the first COVID-19 pandemic wave. The mask finding is buried in the paper, but when using daily mask usage as the intervention measure, we detected only a small beneficial impact of mask wearing, but it was not statistically significant.

RP Rannan-Eliya, N Wijemunige et al. 2021. Increased Intensity Of PCR Testing Reduced COVID-19 Transmission Within Countries During The First Pandemic Wave. Health Affairs.<br /> https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2020.01409

It seems very difficult to detect statistically significant benefits from mask wearing for COVID-19 in global analyses that adequately control for other interventions, suggesting that the benefit is likely to be small at population level. One possible reason is that this is because mask wearing in most contexts is only mandated outside the home, whilst most SARS-CoV-2 transmission occurs inside the home in most countries. Another could be that the transmission blocking is weak in practice owing to problems in how people wear masks, compliance, etc.

On 2021-06-17 09:47:55, user Subhajit Biswas wrote:

Dear Readers,

I am pleased to inform you that the above preprint posted in medRxiv has now been accepted and published by the "Journal of Medical Microbiology". Please see link below:

Title: Archived dengue serum samples produced false-positive results in SARS-CoV-2 lateral flow-based rapid antibody tests

Link: https://www.microbiologyres...

Best wishes to all; stay safe.

Yours sincerely,<br /> Subhajit Biswas (Corresponding Author).

On 2020-09-23 03:00:23, user Amador Goodridge wrote:

Great job of this research. Very important to know what they know. Congrats to the team!

On 2020-05-13 10:14:16, user Shohei Hidaka wrote:

We are posting the predicted evolution of COVID-19 almost everyday.<br /> You can see the visualized result online: <br /> http://www.jaist.ac.jp/proj...<br /> Project Prepidemics

On 2020-05-20 17:49:28, user Christopher Leffler wrote:

Bottom line, how many people does Dr. Ioannidis think will die in the US from this epidemic? If one reads the paper, he proposes that " even under congested circumstances, like cruise ships, aircraft carriers or homeless shelter, the proportion of people infected does not get to exceed 20-45%."<br /> Also, he believes that the infection fatality ratio is: " Infection fatality rates ranged from 0.03% to 0.50% and corrected values ranged from 0.02% to 0.40%."<br /> So, these numbers would give estimates for the United States of:<br /> Low end: 331,000,000 people * 0.2 * 0.0002 = 13,240.<br /> High end: 331,000,000 people * 0.45 * 0.004 = 595,800.<br /> The range is so wide as to provide no useful information. And of course, the pandemic is already at 92,387 deaths in the US, as of May 20, 2020. So we know Ioannidis low end is simply wrong.<br /> We have looked at the mortality in different age groups in New York, among residents and transit workers, and on the Diamond Princess:<br /> https://www.medrxiv.org/con...<br /> Quite early in the pandemic (early April), we showed that if the US followed the course that Italy and Spain had already experienced, we would see 100,000 dead in the US:<br /> https://www.researchgate.ne...<br /> More recently, we showed that if the mortality rates seen in New York MTA / New York State / Diamond Princess were observed nationally, the mortality could be over 600,000, which is the high end for Ioannidis work also:<br /> https://www.researchgate.ne...<br /> So, the bottom line is, that the high end projections from all groups could be quite high indeed. So we will need to be vigilant--wearing masks, protecting the vulnerable, etc. The pandemic is real. To say that it is similar to a typical flu is just plain false. Even Ioannidis own projections do not rule out that this is far worse than the flu. When is the last year the flu killed 92,000 Americans and was on track to kill potentially hundreds of thousands more?

On 2020-05-27 10:39:58, user Victor wrote:

It comes from the CDC.

On 2020-12-09 14:33:35, user Elke Devocht wrote:

After completing formal peer review, this article was recently published by Plos one: <br /> https://journals.plos.org/p...

On 2022-02-15 09:24:31, user Shelly L Miller wrote:

This paper has been peer-reviewed and published here: https://www.nature.com/arti...

On 2025-04-16 20:15:51, user David Gorla wrote:

This is a very interesting article and we thank the authors for this pre-print. There is no discussion about the relevance of dengue especially in Latin America. Living here, I would be delighted to see evidences of a method to prevent dengue outbreaks, that unfortunately is badly hitting this region during the last 20 years without control. I followed with much interest the publications of the WMP and I have to say I have too many doubts to agree with the authors interpretations, with the past articles and the present one. When dengue incidence is included, the main weakness of all WMP articles (except the Yogyakarta one, and that would be another discussion) is that the study designs used in the publications do not allow to make solid interpretations of the results. All of them rely on comparisons of the temporal variation in dengue incidence, that we know is very difficult to explain either in time or in space, and make space for too many interpretations for and against the presented results.

In the present case, the Supplementary Figure S3 Annual dengue incidence in Niterói between 2007 and 2024 (18 months of data) is quite remarkable. Leaving aside incidence between 2019 and 2023 (5 years), identified as “city - wide Wolbachia deployment”, leaves 13 years of data. Dengue incidence in 2024 (after Wolbachia release) is lower than incidence in 6 previous years (2007, 2008, 2011, 2013, 2014 and 2016) and higher than incidence in the other 6 previous years of the period. So, one could say that 2024 was an average year for dengue incidence between 2007 and 2018, period without any Wolbachia influence. Additionally, it is at least an order of magnitude higher than the incidence during the Wolbachia deployment period.

Authors argue that this case is in line with the results of the Colombian case they published some time ago. In the Colombian case data on Wolbachia infection shows a wild variability never quite reaching the magical 60% infection level that should be reached to sustain the introgression (discussion on this case is being prepared to be published elsewhere, although you can have a look at https://davidgorla.substack.com/p/is-dengue-really-controlled-using) "https://davidgorla.substack.com/p/is-dengue-really-controlled-using)") .

Anders et al recognize the incomplete blocking of dengue transmission by Wolbachia infected Aedes aegypti (page 9). It was shown that this is especially the case with DEN1 (one of the most frequent strains affecting Latin America). Adding to that is the evidence (shown on a number of published articles) on loss of Wolbachia infection because of high temperature during summer months, etc.

Wrapping up, the release of Wolbitos show loss of infection during summer months or complete loss of infection, incomplete virus blocking (especially DEN1), and do not show a convincing impact on dengue incidence between treated and untreated areas.

So, if I were to review this article I would reject it based on the lack of data supporting the claims the authors make. And I would also suggest to health authorities of any country considering this dengue control technique to ask for much clearer evidences not only on the results of the WMP claims, but also ask for evidences that the released Wolbitos are not worsening dengue transmission.

On 2021-12-23 23:32:48, user tunneslofreality wrote:

South Africa has a low vaccination rate. Yes, many factors involved in predicting outcome for each geographic location. <br /> Average age in South Africa is 10 years younger than New York, but so far their hospitalization and death data from Omicron is looking similar. South Africa is in Summer, New York is in Winter.

On 2021-04-15 11:01:12, user Ian Viney wrote:

Interesting paper, and although an approximation I think it makes a good point. Having conducted similar studies to reconstruct research income, I share the methodological frustration that funders do not always provide basic financial details for awards (exceptions include UKRI, Wellcome, NIH and many others), institutions do not always provide details of the research projects they secure (exceptions include Kings College London, Edinburgh University and many others), publishers don't capture structured authoritative grant information in their articles (despite the efforts of FundRef etc.), and authenticated information cannot be easily re-used/compiled (despite the efforts of ORCID). As a result there are clearly a lot of investments that your study is missing. One element is the support that Oxford and other UK and international institutions will have provided to the work. The financial details for the grants you have identified will in the main be the amounts awarded by the funder, not the full economic costs of the work. Your FOI might have been an opportunity for you to collect the full cost of each project, and I'm surprised that Oxford didn't comment on this. As most government funded grants are awarded on the basis of 80% of full-economic costs, the institutional contribution may have been ranked third in your analysis. Of course the institutional contribution to the work will be supported from a variety of sources with one major element being the UK funding council grant, so also substantively publicly funded. This would therefore have not have changed your overall conclusion. Some nice context to the public and charity funding for UK health research, with an overall estimate of the contribution from various sectors can be found in the recent UKCRC report at www.hrcsonline.net.

On 2020-09-08 19:17:49, user Jacques des Anges wrote:

Some general data:<br /> in the US for the age group of 45-64 (82million people) there are about 30,000 deaths involved COVID-19, or about 1 in 2667. And about 1 in 2000 for the age group of 55-64. And over 6 million total confirmed cases in the US by sep 5th, or about 1 in 54 people?

While the ensemble level infection and fatality probabilities for individual person to person contact calculated in the study might be useful for epidemiologists, they are not helpful for individuals to do a risk assessment in any way. Even if the results were accurate and the conclusions valid, which is what peer review helps to establish.

The paper give a false sense that the risk is really low when the actual fatality numbers paint a different picture on individual interaction infection probability and overall outcomes. It will cause people to underestimate the probabilities, as is very common; see also: birthday problem. (https://en.m.wikipedia.org/... "https://en.m.wikipedia.org/wiki/Birthday_problem)").

Numbers from CDC as of sep 5 2020. <br /> https://www.cdc.gov/nchs/nv...

On 2021-03-03 00:42:19, user James Gorley, PhD wrote:

In this ambitious study, the authors set out to show histological safety of low intensity FUS. A few key questions should be addressed by the authors. Namely, if the EEG was not usable, how is the claim of "temporal slowing" of one participant justified? Was any statistics or rigor applied to support this claim? Furthermore, two participants are excluded from the analysis, but the data is analyzed later anyway in the psych testing. Interested to see how this manuscript will evolve!

On 2020-02-11 06:09:14, user ) Bill Ford wrote:

They do not know how long it remains potent only they have found 24 days. What is the total? No one yet knows

On 2021-06-19 22:33:09, user Mazda Sabouri wrote:

My initial thought is that the test group might possibly suffer from a bias towards having other non-Covid issues. While those infections detected through the REACT study were chosen at random, those detected through hospital and clinic data, including the asymptomatics, would have had a much higher likelihood of having other non-Covid issues when testing positive for Covid.

I would personally isolate out the infected who were detected through the REACT study and compare their aggregate results to the control group. Even then it's possible that the infected group might still be more prone to having non-Covid issues, but this bias would be much less pronounced.

On 2020-07-18 19:23:30, user Sui Huang wrote:

Thank you for this important work. We had a similar finding in (brief report, purely computational part) that may be of interest: https://www.preprints.org/m...<br /> (medRxiv rejected our submission)

On 2024-08-28 04:54:25, user John Orchard wrote:

My own comments having read many excellent replies from the COVID cautious and COVID stoical community on X in the last 24 hours are summarized at https://x.com/DrJohnOrchard/status/1828598891027333208

On 2021-02-11 19:05:46, user Dr John Tal wrote:

Eagerly awaiting the results of the trials using ciclesonide. Has been shown to slow down SARS CoV2 replication in vitro and seemed to be effective in a couple of case studies. https://jvi.asm.org/content...

On 2022-01-08 03:47:33, user Robyn Chuter wrote:

Also, deaths occurring in people who developed myocarditis due to a breakthrough infection should be distinguished from deaths of people whose myocarditis was not related to breakthrough infection.

This would help identify whether ADE or related phenomena are contributing to myocarditis.

On 2022-01-18 11:36:44, user Krsnendu Knight wrote:

This paper only investigates vaccinated people.<br /> There is a 28 day period prior to vaccination that could give some clues to myocarditis effect of covid-19 infection on unvaccinated people but it would be good to include the myocarditis numbers for those who are never vaccinated and test positive for covid.

On 2020-06-26 17:52:11, user abdullah Alsultan wrote:

Thanks for the great work, few questions regarding your PK study:

1) I think in the Perinel study, they used whole blood conc not serum

2) How did you determine the blood/plasma partitioning ratio? In your study it was 1.64, prior studies was closer to ~5

3) Would be good to add to figure 1, the predicted blood conc bases on the serum conc you observed

Thanks,<br /> Abdullah Alsultan

On 2020-05-01 10:16:00, user mendel wrote:

The specificity trials on page 19 are not normal.<br /> 7 trials show 100%, with N=30,70,1102,300,311,500,99, sum 2412.<br /> The 6 remaining trials:

368/371 = 99.2% (97.7-99.8)<br /> 198/200 = 99.0% (96.4-99.9)<br /> 29/31 = 93.6% (78.6-99.2)<br /> 146/150 = 97.3% (93.3-99.3)<br /> 105/108 = 97.2% (92.1-99.4)<br /> 50/52 = 96.2% (86.8-99.5)

Pooling these, I get 896/912=98.3% (97.2-99.0).

"We use the pooled test performance based on the available information:<br /> Sensitivity: 82.8% (exact binomial 95CI 76.0-88.4%)<br /> Specificity: 99.5% (exact binomial 95CI 99.2-99.7%)"

There is no trial that has exactly 1 false positive. There are 3 <br /> trials that don’t have 99.5% in the 95% CI (4 trials if you include <br /> 1102/1102). There is no trial that falls inside the 99.2-99.7 range (one<br /> straddles it). The specifity range they’re using is an empty space <br /> between the values that the trials are actually at. This is not a normal distribution.

187 samples had loss of smell and taste in <br /> the past 2 months. This is a very specific indicator for Covid-19, ~70% <br /> of patients (well, 33,9–85,6%, depending on the study, e.g. <br /> Mons/Belgium, Heinsberg/Germany) have that, and I don’t think this kind <br /> of nerve affliction has been reported for any other common illness. Yet <br /> only 11% of these samples test positive. For the 59 more recent samples,<br /> it’s 22%.

This looks like the prevalence this study should have measured is <br /> 267/3330 = 8%, and the test failed to pick up on that. It would fail to <br /> pick up on recent infections, because they wouldn’t have seroconverted <br /> (created enough antibodies) yet, and it would fail to pick up on <br /> infections that happened too long ago (because the antibody levels would<br /> have fallen off below the sensitivity of this test). This study really <br /> needed a more sensitive test, like an ELISA, which is actually available<br /> at Standford, and is able to detect much lower levels of antibodies.

This kit has not been validated against people who had the infection a month ago.

The presence of false positives is an indication that cross-reativity <br /> with outher cold viruses exists. If you test a sample with few people <br /> who haven’t had a severe cold recently, which probably includes most <br /> samples taken of people who check into the hospital for elective <br /> surgery, or samples taken in the summer months, you get an optmistic <br /> sensitivity that does not apply to the general population in early <br /> spring.

The WHO Early investigation protocol (Unity protocol) for the <br /> investigation of population prevalence mandates the use of an ELISA <br /> test, or the freezing of samples until a time when such a test becomes <br /> available. The WHO does not endorse the use of lateral flow assays for <br /> this kind of testing.<br /> —-<br /> P.S.: No study that does not measure prevalence in the older <br /> population where the majority of deaths occurs should speak on fatality <br /> rates. This study had 2/167 positives in the age 65+ population, that’s <br /> 0.1-4.3% (95% CI), a 30-fold spread, and hardly a representative sample,<br /> since I don’t expect residents from care homes were able to attend the <br /> drive-through testing.

On 2020-05-01 12:52:41, user Scott Howell wrote:

Any chance that the data and code can be made available? Seems fitting given the topic and re-analysis.

On 2021-03-27 15:05:23, user Rogerblack wrote:

I note the severe concerns raised before the trial about inaccuracy of mental health scales used in this paper are not addressed at all in version 2. To find that comment, click on 'view comments on earlier vesions of this paper'.

In short, mental health scales with physically ill patients risk being akin to asking patients 'do you wobble when you stand up' and concluding that one-leggedness puts you at great risk of low blood pressure.

The measures used confuse 'I can't as I am physically unable to' with 'I cannot as I have anxiety/depression'

Emailed coresponding author and other two leads on 25th, raising these concerns.

On 2020-05-27 15:48:28, user OxImmuno Literature Initiative wrote:

https://www.immunology.ox.a...

On 2021-01-29 13:55:07, user Robyn Wright wrote:

We discussed this preprint in a recent journal club meeting. Very interesting study and thanks for making it available as a preprint! You can read our thoughts on it here if you are interested: http://dalmug.org/mechanica...

Best wishes,<br /> Robyn

On 2021-02-07 11:26:26, user Luzia wrote:

Such precious excellent news! I am delighted that scientists are researching the benefits of natural remedies like propolis. This helps us to take more responsibility for our own health. Lets be grateful to the bees that they collect and process this natural substance and do all we can to protect them.

On 2020-03-28 21:56:25, user V. Cheianov, Esq. wrote:

I believe, z isn't the susceptible population. It is z=(R+I)/N <br /> (recovered + infected)/{total population).

Therefore, calculating total fatalities as fatality rate times z is appropriate. <br /> However, in the situation of exponential growth of z, this way of <br /> evaluating fatalities is exponentially sensitive to the uncertainty <br /> in psi, also to its fluctuation across the population.

On 2020-07-09 08:53:02, user Ahmed wrote:

Applying the logistic growth model to the second period (scenario) separately is unprecedented approach as modeling should include the entire data of a single outbreak. It will be useful to model (scenario 3) which includes the whole period.

On 2021-01-31 18:31:27, user Graeme Ackland wrote:

The statement

"we showed approximately 51% effectiveness of BNT162b2 COVID-19 vaccine against PCR-confirmed SARS-CoV-2 infection 13-24 days"

Is highly misleading. The data suggests more like "15% effectiveness 13-18 days, 85% effectiveness 19-24 days.". The most relevant day is day 21, when the second dose is meant to be given.

So their conclusion is that someone else should be deprived of 85% protective first dose, in order to give an 10% uplift with a second dose.<br /> I find that logic debatable

On 2020-10-22 03:41:09, user koch wrote:

I read your observational study with interest but have some questions about the methods portion. It seemed that exposure / treatment with digoxin was determined by the presence/ absence on the discharge medication list. There is also mention of 2 scripted phone interviews with patients and relatives. What was the role, content, scope, and timing of these interviews ? Were the interviewers “masked “ in terms of awareness of who was exposed/ treated with digoxin? If a patient was discharged on digoxin but stopped before the first interview , how were they categorized ? Conversely if a patient was not discharged on digoxin but was started on it before the first or second interview , how were they categorized ?

On 2021-08-11 15:23:39, user Rene Reeves Brandon wrote:

The study included a cohort of unvaccinated individuals, but only reported on outcomes of individuals fully vaccinated with either of the two vaccines. What did the data on the vaccinated reveal in comparison to the vaccinated regarding previous infection, illness, hospitalization, and death? That data is necessary to share, especially as mandatory vaccines are being discussed in several states.

On 2020-03-24 13:35:07, user Sinai Immunol Review Project wrote:

Summary: Retrospective study of the clinical characteristics of 752 patients with pneumonia infected with SARS-CoV2 , admitted at Chinese PLA General Hospital, Peking Union Medical College Hospital, and affiliated hospitals at Shanghai University of medicine & Health Sciences. This study compares peripheral blood from healthy controls from the same regions in Shanghai and Beijing, and COVID-19 patients to standardize a reference range of lymphocyte counts stratified by age.

Key findings: Lower levels of lymphocyte counts - CD4 and CD8 T cells- correlated with disease severity (T cell counts were significantly lower in critical patients (in intensive care units, ICU) vs non-ICU). Based on 14,117 normal controls in Chinese Han population (ranging in age from 18-86) the authors recommended that reference ranges of people with CD3+ lymphocytes below 900 cells/mm3, CD4+ lymphocytes below 500 cells/mm3, and CD8+ lymphocytes below 300 cells/mm3 be considered high risk of severe COVID-19. However, COVID-19 patients were not stratified by age. This study reported that the levels of D-dimer, C-reactive protein and IL-6 were elevated in COVID-19 pts., indicating clot formation, severe inflammation and cytokine storm, but these parameters were not shown for healthy controls Authors compare data from patients in Shanghai and Beijing with patients in Wuhan, but clinical data from patients in Wuhan are not presented and it is unclear where data from Wuhan were obtained. The authors suggest a correlation between mortality rates and lymphocyte counts when comparing different regions in China, but this claim is not substantiated by data analysis. The authors should revise their title to emphasize disease severity (and not mortality).

Importance: This study sets a threshold to identify patients at risk by analyzing their levels of lymphocytes, which is an easy and fast approach that may stratify individuals that require intensive care Although the study is limited (only counts of lymphocytes are analyzed and not its profile) the data is statistically robust to correlate levels of lymphopenia with disease severity.

By María Casanova-Acebes

On 2021-10-18 20:32:04, user Immede wrote:

Nice, the city of Wuhan is actually out of the zone indicated on their probabilistic maps.

On 2025-10-15 18:43:11, user Ninad Oak wrote:

This paper is now published: https://ascopubs.org/doi/10.1200/JCO-25-01114 .

On 2021-08-04 16:08:46, user Sam Chico wrote:

They compare raw Ct values which are meaningless in qPCR testing and must be correlated with something, e.g. a dilutions series of positive control COVID RNA at the very least. Should really be related back to viral counts using plates and plaques.

This is very lazy research that only shows that vaccinated people can harbor detectable amounts of viral RNA, its quite a reach to say that implications of this data is to say that vaccinated people are infectious. The only way to do that without clearly and reproducibly demonstrating it is looking at the population level data.

Its irresponsible and borderline unethical to publish data like this, the MIQE guidelines made clear what publishable qPCR data should look like. These sorts of publications undermine the impact of properly performed studies with truly actionable data. The authors should know better.

On 2020-04-07 14:53:16, user Quinctius Cincinnatus wrote:

I was glad to see the Imperial College estimate. I'm equally glad to see this work in progress. What happens if only 20% of the population is susceptible to the disease? Diamond Princess had a max of 20% (and no one did a "heat map" of the ship which boggles my mind), Italian hospital workers have a rate of 20% (assuming they have been equally exposed to the virus). We know that the Black Death didn't strike everyone, and it didn't kill everyone it struck. What was the asymptomatic rate used in this study? Did Page four doesn't relay. In Diamond Princess and the Italian village Vo, it appeared to be almost 50%. finally, did anyone look at the potential impact of weather? I suspect Wuhan had more cases than Hong Kong - one reason was the warmer climate. So, as this work in progress continues - it would be good to see those assumptions and look at those variables. .

On 2020-12-11 16:41:00, user Richard Neher wrote:

Review of version 1 of this manuscript -- 2020-12-11:

Kemp and colleagues present a case of persistent SARS-CoV-2 infection and analyze the molecular evolution that unfolded within the host in detail. This case is not dissimilar from two recently described cases (Choi et al (10.1056/NEJMc2031364), Avanzato et al (10.1016/j.cell.2020.10.049)). In contrast to these previous cases, Kemp et al investigate within-host evolution using deep sequencing and trace the frequencies of different variants through time. They characterize three diverged variants with different mutations and deletions in the spike protein, some of which reduce neutralization titers of convalescent plasma in a pseudo-typed lentivirus. Overall, the work in this paper is well performed and it provides convincing evidence of in-vivo antibody escape.

I have a number of suggestions to improve the presentation, strengthen the conclusions, and to remove/tone down parts that might be misleading.

* Fig 2A: The radial tree in Figure 2 is rather unhelpful. The labels are hardly readable and distances between samples are very hard to judge from the radial presentation. A rectangular tree indicating major clades and the different within-host samples would be better.

* Fig 2B & 4B: I think the figure would be improved by changing the scale bar to correspond to one or two mutations (currently the scale bar is given in mutations per site and is roughly 6 mutations in 2B, 2 mutations in 4B). Zero-length branches in the ML trees should be collapsed into polytomies. Bootstrap values on SARS-CoV-2 trees are pretty useless. Better to label the branches with (number of) mutations that fall on the branch in a parsimony or ML reconstruction. This has a one-to-one correspondence to bootstrap values and is more interpretable.

* The purple line in Fig 3B suggests an iSNV at frequency 30% on day one that persists at a frequency around 30% until day 82. This iSNV doesn't seem to be affected by the fixation of other iSNVs at time points 66 or 82 days. Would be good to look into this. It could indicate population structure which would imply parallel evolution. Or it could be an artifact (more likely). Either way, this should be looked at and discussed.

* To understand the rapid shifts in dominating variants better, it would be helpful to include a discussion of their frequencies when they are rare. It makes a difference to the interpretation if the minor variants are present at 10%, 1%, or 0.1%. The reader currently has to piece this together from supplementary table 3 and there are some discrepancies: The S:64G variant seems to be very rare after day 95 (not detected by high coverage Illumina) while the linked S330S is still picked up (at high frequency in low coverage data??). Mutations 200H,240I,258S are missing from supplementary table 3.

* Fig 5 would be more useful on a logscale. Bar charts should be avoided, individual data points need to be shown.

* the description of how evolutionary rates are estimated from within-host data is very short. I would caution against over-interpreting these estimates for two reasons: (i) Phylogenetic estimates are done with consensus sequences and thus ignore minor variation. (ii) rate estimates likely depend a lot on how the within-host variation is rooted and how the root height is constrained. The error of the mean rates (table S2) seems way too small in some cases (1% of the main) calling the entire procedure into question. I would cut this as I don't think this is reliable and it is not central to the paper.

* Similarly, the logistic fit to T39I in ORF7a (Supp Fig 6) is not evidence for selection. I don't see what this figure adds that is not visible in Fig 3. All that Fig S6 shows is that the variant was rare at day one and then bounced around frequency 0.5 between day 30 and 60. There is no reason to fit a logistic and insinuate selection.

* the distances presented in Fig S5 seem rather large (two-fold larger than what I would have guessed from the tree).

* accession numbers for consensus sequences and reads need to be provided.

On 2025-10-18 04:34:17, user CDSL JHSPH wrote:

Hello!

Thank you for sharing this preprint. I found it very interesting, and I believe this is an excellent contribution to improving statistical approaches in antibiotic trial designs. I believe that the comparisons between MCP-Mod and Fractional Polynomials (FP) models provided a very strong case for adopting model-based frameworks to more accurately identify duration-response relationships and estimate the Minimum Effective Duration (MED). I also believe that the demonstration of these methods outperforming traditional pairwise comparisons, particularly in smaller sample sizes, helps highlight the potential to make trials more efficient without compromising reliability and ultimately the patient.

I also appreciate how you clearly discussed the limitations of traditional methods and how there is a need to move towards more model-driven designs. The emphasis on the simulation as a proof-of-concept framework was well-structured and persuasive. I think it would be very interesting to see how these approaches would perform under real clinical data. This is where patient adherence, comorbidities, and variable response rates could truly test the robustness of these models.

I think that in future works, expanding on various factors like dose-spacing, variability in treatment adherence, or non-monotonic response patterns that affect model performance could help further strengthen the clinical applicability of your framework. Overall, I found that this paper sets a solid foundation for refining antibiotic trial methodologies, and also bridges the gap between statistical modeling and real-world trial optimization.

On 2021-05-27 14:01:51, user unscientific science wrote:

Dr.Niaee posted this on the previous version of this article: Hi, I'm Dr.Niaee and I was surprised that even basic data from our RCT is completely mispresented and is WRONG. We had 60 indivisuals in control groups and 120 in intervention groups and even this simple thing is mispresrntated.

You can read all the comments to this article if you click on "View comments on earlier versions of this paper".

On 2021-05-30 16:14:37, user wj wrote:

The amount of misrepresentation of data in this meta-analysis as outlined in below comments reaches the level of scientific misconduct, given that the Niaee RR was corrected but the associated conclusion was not. I have faith that peer reviewers will reject based on this glaring falsity if the authors do not retract beforehand.

On 2024-04-26 22:22:12, user Sailing Pelagia wrote:

Now published in The Canadian Journal of Speech-Language Pathology and Audiology.<br /> https://www.cjslpa.ca/detai...

On 2020-04-21 23:29:37, user Sinai Immunol Review Project wrote:

Title: Factors associated with prolonged viral shedding and impact of Lopinavir/Ritonavir treatment in patients with SARS-CoV-2 infection?<br /> Keywords: retrospective study – lopinavir/ritonavir – viral shedding

Main findings:<br /> The aim of this retrospective study is to assess the potential impact of earlier administration of lopinavir/ritonavir (LPV/r) treatment on the duration of viral shedding in hospitalized non-critically ill patients with SARS-CoV-2. <br /> The analysis shows that administration of LPV/r treatment reduced the duration of viral shedding (22 vs 28.5 days). Additionally, if the treatment was started within 10 days of symptoms onset, an even shorter duration of virus shedding was observed compared to patients that started treatment after 10 days of symptoms s onset (19 vs 27.5 days). Indeed, patients that started LPV/r treatment late did not have a significant median duration of viral shedding compared to the control group (27.5 vs 28.5 days). Old age and lack of LPV/r administration independently associated with prolonged viral shedding in this cohort of patients.

Limitations:<br /> In this non-randomized study, the group not receiving LPV/r had a lower proportion of severe and critical cases (14.3% vs 32.1%) and a lower proportion of patients also receiving corticosteroid therapy and antibiotics, which can make the results difficult to interpret.<br /> The endpoint of the study is the end of viral shedding (when the swab test comes back negative), not a clinical amelioration. The correlation between viral shedding and clinical state needs to be further assessed to confirm that early administration of LPV/r could be used in treating COVID-19 patients.

Relevance:<br /> Lopinavir/ritonavir combination has been previously shown to be efficient in treating SARS [1,2]. While this article raises an important point of early administration of LPV/r being necessary to have an effect, the study is retrospective, contains several sources of bias and does not assess symptom improvement of patients. A previously published randomized controlled trial including 200 severe COVID-19 patients did not see a positive effect of LPV/r administration [3], and treatment was discontinued in 13.8% of the patients due to adverse events. Similarly, another small randomized trial did not note a significant effect of LPV/r treatment [4] in mild/moderate patients. A consequent European clinical trial, “Discovery”, including among others LPV/r treatment is under way and may provide conclusive evidence on the effect and timing of LPV/r treatment on treating COVID-19.

1. Treatment of severe acute respiratory syndrome with lopinavir/ritonavir: a multicentre retrospective matched cohort study. - PubMed - NCBI. https://www-ncbi-nlm-nih-go.... Accessed March 30, 2020.
2. Role of lopinavir/ritonavir in the treatment of SARS: initial virological and clinical findings. - PubMed - NCBI. https://www-ncbi-nlm-nih-go.... Accessed March 30, 2020.
3. Cao B, Wang Y, Wen D, et al. A Trial of Lopinavir–Ritonavir in Adults Hospitalized with Severe Covid-19. New England Journal of Medicine. March 2020. doi:10.1056/NEJMoa2001282
4. Li Y, Xie Z, Lin W, et al. An Exploratory Randomized, Controlled Study on the Efficacy and Safety of Lopinavir/Ritonavir or Arbidol Treating Adult Patients Hospitalized with Mild/Moderate COVID-19 (ELACOI). Infectious Diseases (except HIV/AIDS); 2020. doi:10.1101/2020.03.19.20038984

Reviewed by Emma Risson as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn School of Medicine, Mount Sinai.

On 2022-01-04 00:38:33, user Leicesterboy wrote:

This study looks at 29,000 Omicron patients, but the authors seem unable to find the courage to look at the statistics for “risk of death” independently, preferring to conflate it with other outcomes, like ICU or hospitalizations. Why? Isn’t that the outcome most of us are interested in and fear? By the way, I’d also be keen to see people be more definitive about the word “Hospitalizations” with COVID. We now know that hospitalizations arising from COVID are conflated with Hospitalizations with COVID, meaning those admitted due to an ankle sprain and discharged quickly are included in the “Hospitalizations” numbers. Clearly this is meaningless if the purpose of the analysis is to look at risk from COVID. The fact that someone with a sprained ankle had asymptomatic COVID has absolutely no interest for me. It Carrie’s zero information about health risk and potential outcomes for me.

On 2020-04-24 17:27:51, user dak wrote:

How do you know that this is the virus and not RNA fragments produced in fighting the infections? If it was a whole virus, would you not expect the N1-N3 PCR amplicons to parallel each other? You could argue about the stability of RNA in waste water under conditions X, if that is known for this specific RNA fragment and the conditions X expected.

On 2020-05-02 20:41:46, user wubba lubba dub dub wrote:

1. The CDC numbers may not be the actual ones. 2. Does not seem to count the fact that the number of testing has increased.

On 2020-07-23 04:43:58, user DocWhy wrote:

Nice study. Thank you. I would like to see numbers or percent distribution of cases by cluster. This would be very helpful.

On 2021-01-19 19:01:38, user Filip Hrivnák wrote:

That voluntary. In general, we are losing the last signs of the rule of law

On 2020-05-22 22:54:11, user Snap wrote:

An updated version of this paper with 8 countries can be found here https://www.medrxiv.org/con...