On 2020-07-24 04:38:59, user Dr Prachi Sinkar wrote:

How are you sure it was not against COVID? if you are negative when you are tested for COVID but exposed to it before or after - still Ab against COVID possible?

On 2021-12-16 16:22:57, user itellu3times wrote:

While I understand the place for these calculations given the social and government actions around the world, even the necessity for someone to do it, and it comes out on the correct side of things, I must point out that it is also socially deranged and mathematically extremely vague. "The unvaccinated" do not even exist as such in US, EU, UK, and other places where the pandemic has run now for two years, almost all will now have contracted and recovered from COVID, have natural immunity, and be no more at risk of getting or giving COVID than any of "the vaccinated". I would like to see studies of just where these "unvaccinated" patients today are even coming from - are they all immunocompromised, or vitamin D compromised, or are the figures cooked by authorities who use biased procedures "for reasons of public health", to bias the results? We know "the vaccinated" transmit the virus too, is that supposed to comprise the "base rate" here? How about reversing this, what if the vaccinated are excluded, what's the NNT then? But, whatever the calculation here, forward or reverse, the odds calculated are for one event, and when there are N such events even small odds are enough to propagate. Now, perhaps this factor is already comprised in this "base rate", I guess it is, but then the real odds for a single event may be much lower. You are then talking about socially exclusionary processes that even the article states are really for reasons of coercion, for event numbers that are truly tiny. Is this rational? Oh, and the SAR numbers must be considered plus or minus 80%, instances of any type event may be vastly better, or worse. Very hard to use such things to base any serious decisions on at all. Though, in science, I guess someone has to give it a look, as seems to have been done reasonably here.

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 2021-09-16 08:25:05, user Dr Siona Bastable wrote:

Overlap with effects of ADHD symptomology is striking....

On 2020-07-27 06:46:38, user OxImmuno Literature Initiative wrote:

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

On 2020-07-27 13:58:22, user Rosemary TATE wrote:

Excellent and interesting paper. However, although you say you adhered to the relevant EQUATOR (TRIPOD) guidelines I note that you have not uploaded the checklist. Very few people seem to do this although they tick the box that they have. I'm wondering why? Can you enlighten me?

On 2020-12-30 22:01:49, user Henry wrote:

" translated to a rise of 21.1 nmol/L of 25OHD in the UK Biobank population, a rise that is comparable to what can be achieved with vitamin D supplementation, especially in short courses[38]."

A 21nmol/liter serum raise is not much. That is what you get when you supplement too little vitamin D (400 iu / day for short courses).

Did you compare 150 nmol / L and higer to 30 nmol / L? I would say someone has optimal vitamin d at 150 nmol / L.

On 2021-09-19 05:15:21, user Les Smith wrote:

The background rates between 2017 and 2019 are not a valid basis for comparison. The rates of GBS, Bell's Palsy, Neuralgic Amyotrophy, and other such disorders have been significantly suppressed during the pandemic by efforts such as masking and isolation.

On 2020-08-08 00:52:19, user Cyraxote wrote:

The covidestim site shows 12 rows of 4 states. That's 48 states. Maryland is one of the missing ones, but I don't know the other.

On 2021-01-20 15:30:41, user Zuzana Kollarova wrote:

This statement is NOT TRUE and the citizens of Slovakia have no idea they are a part of some medical research:<br /> "All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived - Yes"

We have been all forced to this by a number of restrictions and consequences presented by the prime minister and government prior the testing and they let the "choice" to us. If we wouldn´t take part on that testing we couldn´t go to work, to any store, bank, post office etc.. Only basic needs could by fulfilled like grocery shopping, pharmacy etc. Healthy people who refused to take part on this had to stay at home in quarantine like they were infected and could go outside without the risk of getting a fine, if a police would control them randomly on the streets. This lasted 14 days.

They used army, our president found out just from the papers and not officially. She has been called a traitor by the prime minister just one day before the mass operation should start, when she asked for a really voluntary participation for the citizens.

The testing has been done by anonymous, also not always professional medical staff, without knowing their names and place of work.

Those blue papers (test result confirmation) do not contain the necessary legal requirements to be called a "certificate" officially by the law.

And now, we are in the middle of 2nd mass "screening" now, since Jan 18 2021 during the winter, even though the scientist didn´t recommend it at all in current situation.

And again- no one is collecting our written and signed consent. From Jan 27 2021 there will be again 2 groups of people - the "blue" ones and the rest of us. The country will be then split into two half by the results and the worse half of the country has to undergo this procedure 1-2 times again until the Feb 07 2021 and until our prime minister will be satisfied with the results...

On 2021-02-07 17:39:47, user Martin Gažák wrote:

Dear all, regarding the statement "All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived." is a lie. I as a citizen of Slovak Republic participated already 4 times on AG testing, every time under the threat of ban of movement, ban of work, yesterday because without my negative test my child would not be accpeted at school. I was never informed that I am participating on any sort of research. I did not sign anything. I consider the above statement as misleading and unethical, especially because the authors are government officials. Best regards, Martin Gazak

On 2021-08-21 22:44:12, user Ands Hofs wrote:

Can we please introduce more calibrated PCR that measures the mucosa DNA count and gives out<br /> Viral Load = viral units / mucosa DNA units?<br /> Only then the force and technique of swabbing is not changing the resulting viral load wildly.

Even RTLAMP.org is able to do this, open science test, might be liked by some in the comments here, done in a parallel test, and Boston Children's Hospital did a very good job showing children sometimes have 10x higher calibrated viral loads, which has to be treated early with determination in nasopharynx and mouth, like with xylitol + Grape Seed Extract nasal spray, puff and breathe in a bit to protect vocal chords, upper trachea, whole nasopharynx. Vulnerables: add azelastine as pre-spray dito. Report on CARVIN (11.8.2021) shows excellent efficacy.

If you want "life" reproduction rate, you have to train a dog, see scent dog identification of samples .. covid. Work of TiHo, small animals university Hannover. Built a training device they call scent learning box, like a game for the dog. In one week it is on 95% congruency of PCR, but better: 4 days before infectiousness, and quasi live. It doesn't over-diagnose and does diagnose viral replication. <br /> There is a group that built a speech interface for a dog. This would enable to train the dog on many illnesses, differentiate flu from covid, and even predict severe case, as it can smell susceptibility to autoantibodies (or in another picture: prevalence of MCAS, see Prof. Afrin on Covid. Having deep implications on therapy and prevention. The tricky part is to diagnose MCAS with its 200 mostly congruent symptoms to Post-Covid. So obviously related except scarred tissue of course.)

Even better: build an electronic nose as sensitive as a dog's nose on some marker molecules for covid and a tensor flow neural network in a mobile phone to read it out. The do progress with mice conk cell detecting proteins they use as film on a chip based electronic nose.

On 2021-08-24 01:52:14, user Raihan Farhad wrote:

Please answer the following, in the interest of academic integrity: <br /> 1. What is the effectiveness of masks used in your model? What number did you use? What type of mask? Worn in what way. I can only assume, given the absence of any experimental data regarding un-regulated masks stopping Covid aerosols, you either assumed the effectiveness of a mask or used someone else's assumption. Please divulge.

1. What is the assumption you made about % of kids having been exposed already? Covid has been around for a while now. If you assumed 0 previous exposure, that is unrealistic, but please state so clearly. If you assumed any other number, please explain how you came to that number and state that number.

2. What is the duration of infectiousness assumed in your model. According to science, someone infected is infectious for about 5 days. After that, even if he dies, he is not infectious. Please explain the temporal nature of infectiousness assumed in your model.

3. Please make your entire model / simulation software (all code) and all parameters, assumptions public.

On 2020-08-18 17:33:47, user Rodolfo Rothlin MD wrote:

Dear Dr. Turgeon,<br /> Thank you for your commentary and your interest in our manuscript. Please, find my answers below.

1. Please provide the rationale for selecting July 31 as the date for interim analysis. Please also provide details regarding this interim analysis, including pre-specified stopping rules, who had access to the data. Although this manuscript is labeled as a "preliminary report", it would be valuable for the authors to explicitly state whether this trial is ongoing, and whether any changes to the conduct of the trial were made based on this interim analysis.

The rationale for selecting July 31st was made for several reasons: First, we assumed that by that date we would have between 50 and 100 patients. Although our estimated sample size was 390 (which we rounded to 400), it is worth noticing that this number accounts for a scale factor on both the mean and the variability estimations; without these factors, the estimated sample size was 52 patients total, and with only a variability factor of 2 the total estimated number was 100. Therefore, we evaluated that July 31st was an appropriate time to make the first interim analysis. Second, as you may know, Argentina was expected to be peaking around that time. And it actually seems that it may be doing it right now. So, a second reason for that date was our prediction that if the results were valuable it could be a useful information for our health authorities. The trial is still ongoing and a second interim analysis will be carried out at 140 patients.<br /> 2. In version 1 of the article on this site, the Methods section had a sentence that stated "No concealment mechanism was implemented". This was subsequently removed in version 2 yesterday. Please clarify what is meant by this. Did the authors mean to imply that allocation concealment was not performed, or was this an erroneous statement intended to describe the unblinded nature of the study? Please also describe the process for treatment allocation and how allocation concealment was maintained.

The sentence you refer to was removed because it was inaccurate. The problem emanated from the fact that our protocol did not foresaw a concealment mechanism. However, during the conduct of the trial, although no mechanism like closed envelopes with randomization was used, on site enrollment was made by an investigator and randomization was made by a second investigator who was unaware of the clinical characteristics of the participant. We are confident that no bias towards the control group was present as reflected by data on table 1 of our manuscript.<br /> 3. The authors describe a change in the primary outcome in terms of timing of CRP measurements. However, I note that the clinicaltrials.gov summary of this trial previously had an entirely different outcome as the primary outcome, with CRP only described as an exploratory/tertiary outcome. The authors should describe the timing and rationale for switching the outcome from a clinical one (need for supplemental oxygen in the first 15 days post-randomization) to the inflammatory biomarker CRP.

As you might have read in the methods section of our manuscript, data from our trial was uploaded by a third party. Unfortunately, endpoints from a working version of the protocol were submitted. This was corrected as soon as we noticed it.<br /> 4. Despite changing the timing of CRP measurements, data on this modified primary outcome of CRP was missing in a large proportion of patients at day 5, and in the majority of patients at day 8. Further details should be provided regarding the reason for missing data, how this was handled in their analyses, and how this should temper conclusions.

Data missing from days 5 and 8 were related to several factors. Some patients were discharged before day 5 and before day 8. Others were lost at day 5 for logistical reasons. <br /> No imputations were done to account for these values.<br /> 5. Finally, performing an interim analysis and disseminating their results in the midst of an open-label trial with subjective endpoints can pose challenges to maintaining impartiality. The authors should describe how they will mitigate potential allocation, performance, and detection and attrition bias during the remainder of the trial.

We disagree with CPR measurements being subjective<br /> Again, thank you for helping us clarify these points.<br /> All the best,<br /> Rodolfo Rothlin

On 2021-01-30 21:28:00, user Sharif Ismail wrote:

This article has now been published in PLoS ONE.

It can be found here: https://journals.plos.org/p...

On 2021-02-03 00:57:08, user kdrl nakle wrote:

Internationa Univerlsity?<br /> C'mon, nobody proofreading the paper?

On 2021-10-01 18:41:38, user Paul Tupper wrote:

Thanks for this very valuable study. Do you have plans to do a similar one for younger age groups?

On 2020-08-24 11:06:25, user Atif Habib wrote:

An excellent paper which provided the importance of short birth intervals and the associated factors in the context of Pakistan. The results indicate that lack of contraception and illiteracy significantly contribute to the problem however it is pretty evident that priority should be given to modern contraception which is comparatively a low hanging fruit in comparison to averting illiteracy.

On 2021-06-25 00:23:11, user Otheus wrote:

While I would like to believe the results of this study, the details and the presentation of their numbers leaves much to be desired for the numerically astute/obsessed. A CI of "0 to infinity" means someone has not really done the statistics right at all. In the online preprint of the article, which may be an older version, the article cites -- somewhat deceptively -- provides the percentages in terms of the number of infections in one group with respect to the number of infections in another group. So, 99.3% of the infections were from people not previously infected and not vaccinated and 0.7% of infections were from vaccinated group. The problem with doing it that way is that you have a much larger population of people not infected and not vaccinated than the other sub-populations.

Imagine you have two drawers of socks, and in each drawer, there is the same ratio of white socks to red socks -- let's say 1 red sock for every 9 white socks, ie, 10% red. You then pull out 10 socks from each drawer. From the first drawer, you pull 5 red socks and 5 white socks, and from the second drawer, you pull 1 red sock. It would seem the first drawer had a greater percentage of red socks. In fact, the first drawer had 500 socks and therefore 50 red socks, while the second drawer had exactly 10 socks and only 1 red sock. The probability of this happening in each case is about the same. The number of red socks to socks drawn from the first drawer is 50%, but for the second drawer, it is 10%. But if you look carefully, the first drawer has 5x the number of red socks as the second. The ratio of red socks drawn to total *red* socks in that drawer is 10% in both cases.

At any rate, from the math given, I calculate that the rate of those got sick among those who were not vaccinated and did not have previous infection was about 10%, while the rate of those who got sick among those who either had been vaccinated or had a previous infection was about 1.2%. Since 0 people with a previous infection reported getting sick, we get 0%. What's significant here is that the population of those unvaccinated and not infected was 10x higher than that last group.

An important sentence from the paper sticks out: "Not one of the 2579 previously infected subjects had a SARS-CoV-2 infection, including 1359 who remained unvaccinated throughout the duration of the study." Those numbers appear high enough with respect to the lower bound of the infection rate (1.2%) to have enough statistical power. You'd expect to find at least 31 cases for the null hypothesis. It seems quite improbable to get 0 results unless previous infections provide very strong protection.

On 2020-08-26 13:53:31, user Melimelo wrote:

This is amazing. Well done! Can you share pictures of the equipment you made?

On 2020-05-03 17:48:53, user Sinai Immunol Review Project wrote:

SUCCESSFUL MANUFACTURING OF CLINICAL-GRADE SARS-COV-2 SPECIFIC T CELLS FOR ADOPTIVE CELL THERAPY

Leung Wing et al.; medRxiv 2020.04.24.20077487; https://doi.org/10.1101/202.... 20077487

Keywords

• SARS-CoV-2 specific T cells

• Adoptive T cell transfer

• COVID-19

Main findings

In this preprint, Leung et al. report the isolation of SARS-CoV-2-specific T cells from two convalescent COVID-19 donors (n=1 mild, n=1 severe; both Chinese Singapore residents), using Miltenyi Biotec’s fully automated CliniMACS Cytokine Capture System: convalescent donor PBMCs were stimulated with MHC class I and class II peptide pools, covering immunodominant sequences of the SARS-CoV-2 S protein as well as the complete N and M proteins; next, PBMCs were labeled with a bi-specific antibody against human CD45, a common leukocyte marker expressed on white blood cells, as well as against human IFN-?, capturing T cell-secreted IFN-? in response to stimulation with SARS-CoV-2 peptides. Post stimulation, IFN-?+ CD45+ cells were identified by a mouse anti-human IFN-? antibody, coupled to ferromagnetic dextrane microbeads, and magnetically labeled cells were subsequently isolated by positive immunomagnetic cell separation. Enriched IFN-?+ CD45+ cells were mostly T cells (58%-71%; CD4>CD8), followed by smaller fractions of B (25-38%) and NK cells (4%). Up to 74% of T cells were found to be IFN-?+, and 17-22% of T cells expressed the cytotoxic effector marker CD56. Very limited phenotyping based on CD62L and CD45RO expression identified the majority of enriched CD4 and CD8 T subsets as effector memory T cells. TCR spectratyping of enriched T cells further revealed an oligoclonal TCR ß distribution (vs. a polyclonal distribution pre-enrichment), with increased representation of Vß3, Vß16 and Vß17. Based on limited assumptions about HLA phenotype frequencies as well as estimated haplotype sharing among Chinese Singaporeans, the authors suggest that these enriched virus-specific T cells could be used for adoptive cell therapy in severe COVID-19 patients.

Limitations

This preprint reports the technical adaptation of a previously described approach to isolate virus-specific T cells for targeted therapy in hematopoietic stem cell transplant recipients (reviewed by Houghtelin A et al.: https://www.ncbi.nlm.nih.go... "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5641550/pdf/fimmu-08-01272.pdf)") to PBMCs obtained from two convalescent COVID-19 patients. However, not only is the title of this preprint misleading - no adoptive cell transfer was performed -, but this study also lacks relevant information - among others - on technical details such as the respective S epitopes studied, on the precise identification of immune cell subsets (e.g. NK cells: CD56+ CD3-?), data pertaining to technical stimulation controls (positive/negative controls used for assay validation and potentially gating strategies), as well as on the percentage of live enriched IFN-?+ CD45+ cells. Generally, a more stringent phenotypical and functional characterization (including coexpression data of CD56 and IFN-? as well as activation, effector, proliferation and other markers) would be advisable. Similarly, in its current context, the TCR spectratyping performed here remains of limited relevance. Most importantly, though, as noted by the authors themselves, this study is substantially impaired based on the inclusion of only two convalescent donors from a relatively homogenous genetic population as well as by the lack of any potential recipient data. In related terms, clinical criteria, implications and potential perils of partially HLA-matched cell transfers are generally not adequately addressed by this study and even less so in the novel COVID-19 context.

Significance

Adoptive cell therapy with virus-specific T cells from partially HLA-matched third-party donors into immunocompromised recipients post hematopoietic stem cell transplantation has been successfully performed in the past (cf. https://www.ncbi.nlm.nih.go... https://www.jci.org/article... "https://www.jci.org/articles/view/121127)"). However, whether this approach might be clinically feasible for COVID-19 therapy remains unknown. Therefore, larger, more extensive studies including heterogeneous patient populations are needed to assess and balance potential risks vs. outcome in the new context of COVID-19.

This review was undertaken by V. van der Heide as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.

On 2020-05-05 03:43:04, user Sinai Immunol Review Project wrote:

A possible role of immunopathogenesis in COVID-19 progression

Anft M., Paniskaki K, Blazquez-Navarro A t al.; medRxiv 2020.04.28.20083089; https://doi.org/10.1101/202...

Keywords

• SARS-CoV-2 spike protein-specific T cells

• COVID-19

• adaptive immunity

Main findings

In this preprint, 53 hospitalized COVID-19 patients, enrolled in a prospective study at a tertiary care center in Germany, were assigned to moderate (n=21; light pneumonia), severe (n=18; fever or respiratory tract infection with respiratory rate >30/min, severe dyspnea, or resting SpO2 <90%), and critical subgroups (n=14; ARDS, sepsis, or septic shock) according to clinical disease. Moderately and severely ill patients with a PCR-confirmed diagnosis were recruited within four days of clinical onset, whereas critically ill patients were enrolled on average within 14 days of diagnosis on admission to ICU. To account for the overall longer hospital stay in ICU cases prior to inclusion, repeated blood samples were obtained from moderately and severely ill donors within eight days post recruitment. For 10 out of 14 ICU patients, no follow up blood samples were collected. At recruitment as well as on follow-up, circulating lymphocyte counts were below reference range in the majority of enrolled COVID-19 patients. Relative frequencies were significantly reduced in critically vs. moderately, but not vs. severely ill individuals, with substantially lower NK as well as CD8 T cells counts, and a concomitant increase of the CD4:CD8 T cell ratio in ICU patients. Basic phenotypic and immune cell subset analysis by flow cytometry detected lower frequencies of central memory CD4 T cells as well as reduced terminally differentiated CD8 Temra cells in critical COVID-19. Moreover, a decrease in activated HLA-DR+ CD4 and CD8 T cells as well as in cytolytic CD57+ CD8 T cells was observed in critical vs. severe/moderate disease. Similarly, frequencies of CD11a+ CD4 and CD8 T cells as well as CD28+ CD4 T cells were lower in critically ill donors, indicating a general loss of activated bulk T cells in this subgroup. In addition, a reduction of both marginal and transitional CD19+ B cells was seen in patients with severe and critical symptoms. Of note, on follow-up, recovering severe COVID-19 patients showed an increase in bulk T cell numbers with an activated phenotype. Importantly, SARS-CoV-2 spike (S)-protein-specific CD4 and CD8 T cells, identified following stimulation of PBMCs with 15-mer overlapping S protein peptide pools by flow-cytometric detection of intracellular CD154 and CD137, respectively, were found in the majority of patients in all COVID-19 subgroups at the time of recruitment and further increased in most subjects by the time of follow-up (antiviral CD4 >> CD8 T cells). Most notably, frequencies of both antiviral CD4 and CD8 T cells were substantially higher in critically ill patients, and virus specific CD4 and CD8 T cells in both critically and severely ill subgroups were shown to produce more pro-inflammatory Th1 cytokines (TNFa, IFNg, IL-2) and the effector molecule GzmB, respectively, suggesting an overall increased magnitude of virus-specific T cell inflammation in the context of more severe disease courses. Furthermore, frequencies of antiviral CD4 T cells correlated moderately with anti-S-protein IgG levels across all patient groups.

Limitations

In general, this is a well executed study and most of the observations reported here pertaining to overall reduced bulk T cell frequencies (along with lower NK and other immune cell counts) as well as diminished numbers of T cells with an activated phenotype in ICU vs. non ICU COVID-19 corroborate findings in several previous publications and preprints (cf. https://www.jci.org/article... https://academic.oup.com/ji... https://www.nature.com/arti... https://www.medrxiv.org/con... https://www.medrxiv.org/con... "https://www.medrxiv.org/content/10.1101/2020.04.17.20061440v1.full.pdf)"). Notably, in contrast to many previous reports, the prospective study by Anft et al. enrolled a relatively larger number of COVID-19 patients of variable clinical disease (with the exception of mild cases). However, there are a few weaknesses that should be addressed. Most importantly, the choice of statistical tests applied should be carefully revised: e.g. comparison of more than two groups, as seems to be the case for most of the figures, requires ANOVA testing, which should ideally be followed by post-hoc testing (despite the somewhat confusing statement that this was conceived as an exploratory study). Given the overall limited case numbers per clinical subgroup, trends even though they might not reach statistical significance are equally important. Similarly, some statements are overgeneralized and should be adjusted based on the actual data shown (e.g. the authors continue to refer to gradual reductions of activated T cell subset numbers in moderately vs. severely vs. critically ill patients, but for the majority of data shown substantial differences are apparent only in ICU vs. non-ICU patients). Moreover, it would be helpful to include representative FACS plots in addition to explanatory gating strategies provided in the supplemental document. There are also several inconsistencies regarding the order of data presented here (e.g. in the main manuscript, Fig S5 is chronological referred to before Fig S4) as well as pertaining to relevant technical details (according to both the main manuscript and the gating strategy in Figure S5, virus-specific CD4 T cells were identified by CD154 expression; however, in figure legend S5 virus-specific CD4 T cells are defined as CD4+ CD154+ CD137+). Additionally, from a technical point of view, it is somewhat intriguing that the percentages of virus-specific T cells identified by expression of CD154 and CD137, respectively, following peptide simulation seem to differ substantially from frequencies of CD154+ or CD137+ INFg+ virus-specific T cells. Assuming a somewhat lower extent of cellular exhaustion in the moderate COVID-19 group, one would expect these cell subsets to mostly overlap/match in frequencies, therefore suggesting slight overestimation of actual virus-specific T cell numbers. In this context, inclusion of positive controls, such as CMV pp65 peptide stimulation of PBMCs from CMV seropositive donors, in addition to the already included negative controls would also be helpful. Moreover, in view of the observation that virus-specific T cells were found to be increased in critically ill ICU over non-ICU patients, a more stringent characterization of these patients as well as assessment of potential associations with clinical characteristics such as mechanical ventilation or death would add further impact to the findings described here. Finally, this study is limited to anti-S protein specific T cells. However, evaluation of N and also M-protein specific T cell responses are likely of great interest as well based on current knowledge about persistent M-protein specific memory CD8 T cells following SARS-CoV-1 infection (cf. https://www.microbiologyres... "https://www.microbiologyresearch.org/content/journal/jgv/10.1099/vir.0.82839-0)").

Significance

In addition to reduced frequencies of activated bulk T cell numbers, the authors report an enhanced virus-specific T cell response against S protein epitopes in critically ill COVID-19 patients compared to severely and moderately ill individuals, which correlated with anti-S protein antibody titers (also cf. Ni et al.: https://doi.org/10.1016/j.i... "https://doi.org/10.1016/j.immuni.2020.04.023)"). This is an important observation that mirrors previous data about SARS-CoV-1 (cf. Ka-fai Li C et al.: https://www.jimmunol.org/co... "https://www.jimmunol.org/content/jimmunol/181/8/5490.full.pdf)"). Furthermore, in accordance with a recent preprint by Weiskopf et al. (https://www.medrxiv.org/con... "https://www.medrxiv.org/content/10.1101/2020.04.11.20062349v1.full.pdf)"), virus-specific CD4 T cells were found to increase in most patients over time regardless of clinical disease, whereas antiviral CD8 T cell kinetics seemed slightly less pronounced. Moreover, in the majority of moderately and severely ill cases, virus-specific T cells against the S protein could be detected early on - on average within 4 days of symptom onset. Longitudinal studies including larger numbers of COVID-19 patients across all clinical subgroups are therefore needed to further evaluate the potential impact of this observation, in particular in the context of previously described pre-existing memory T cells cross-reactive against human endemic coronaviruses (cf. https://www.medrxiv.org/con... https://journals.sagepub.co... "https://journals.sagepub.com/doi/pdf/10.1177/039463200501800312)").

This review was undertaken by V. van der Heide as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.

On 2021-02-19 21:51:45, user Michael Verstraeten wrote:

Did you consider to imply Flaxman e.a., Estimating the effects of non-pharmaceutical interventions on Covid-19 in Europe, Nature, 584, 257 - 261 in your study, or didn't it meet the inclusion criteria?

On 2021-07-08 05:59:35, user Dr.G.R.Soni wrote:

Comments on preprint medRxiv publication entitled "Efficacy, safety and lot to lot immunogenicity of an inactivated SARS-CoV-2 vaccine (BBV152) : A double blind, randomised, controlled phase 3 trials" reg.

This is regarding indigenously developed inactivated SARS-CoV-2 vaccine by M/s Bharat Biotech, Hyderabad by using vaccine strain NIV-2020-770 containing D614G mutation. The conventional vaccine consists of 0.5 ml volume having 0.6 ug of virus antigen and results of phase 3 clinical trial available on preprint of medRxiv publisher suggest that the vaccine is of good quality if not best//excellent and the vaccine in my opinion can be used by many under developed and developing countries. However, following are the further comments:

1. Results claimed for vaccine efficacy against severe symptomatic, asymptomatic and delta variant like 93.4%(95% confidence interval CI 57.1-99.8), 63.35( CI 29.0-82.4) and 65.2% (CI 33.1-83.0) respectively are statistically highly insignificant because of lesser precision and wide confidence interval. Even over all vaccine efficacy reported to be 77.8% (CI 65.2- 86.4) is not highly significant. This may be due to known and unknown variations as well as lesser number of participants involved in clinical trials. Of course the study was designed to obtain a two sided 95% CI for vaccine efficacy with lower limit greater or equal to 30% but it is only applicable when vaccines of high efficacy are not available. Whereas now fact is that the efficacy of Moderna, Pfizer, Johnson & Johnson, Sputnik etc. vaccines has been reported to be more than 90.0% with better precisions.

2. Again vaccine efficacy reported for elderly patients viz. 67.8% (CI 8.0- 90.0) shows very poor precision and widest CI means high uncertainty and least confidence in data.

3. No vaccine efficacy data submitted after first dose of vaccine administration and reason furnished by the sponsor is because of low number of Covid-19 cases reported and this is not very convincing. Comparison of vaccine efficacy between two doses is important to know the progress of efficacy.

4. GMT was reported to be higher i.e.194.3 ( CI 134.4-280.9) in vaccinees who were seropositive for SARS-CoV-2 IgG at base line than in those who were seronegative 118.0 (104.0-134.0). The difference is not even two fold whereas in other studies including Pfizer more than 4-6 fold increase in immune response has been reported even after single dose of vaccine under such conditions. This may be due to killed nature of present vaccine which is in general less immunogenic than live vector and m-RNA based Covid-19 vaccines.

5. The immune response of three vaccine lots of vaccine in terms of GMT 50 is 130, 121.2 and 125.4 Vs 13.7 in placebo which seems to be optimum but much higher immune response has been reported in other internationally approved vaccines.Why the IgG GMT titre after two doses of vaccine studied by ELISA has not been reported separately for each lot of vaccine rather overall titre against S1 protein, N protein and RBD has been reported i.e. 9742 EU/ml, 4161 EU/ml and 4124 EU/ml respectively? Since RBD is part of S1 protein therefore S1 titre includes RBD titre also. It means ELISA IgG antibody titre of these viral proteins are roughly equal hence this needs clarification. Anyway unless the titre of these neutralizing and ELISA IgG antibodies are compared with sera of asymptomatic, symptomatic and severely recovered covid-19 patients or immune sera of WHO, US, EU approved vaccines available in market it is very difficult to say whether the immunogenicity of present vaccine is at par or not with these standards?

6. The lesser efficacy of present vaccine than other approved vaccines by the US, WHO and EU may also be due to lack of T cell mediated cytotoxicity response; this is why this response is not measured in the present study.

7. When Covid-19 disease is known to affect men and women differently so separate clinical trial data are required to be submitted by the sponsors. However in the present study no marked difference in GMT,s for neutralizing antibodies at day 56 was found when assessed based on age and gender. This is very surprising and difficult to believe because age definitely and gender also are known to affect the immune response of any viral vaccine.

8. Reasons for contracting Covid-19 disease by some vaccinees are to be given specially when immune-compromised and immunosuppressed etc. patients have already been excluded from the study.

9. As per WHO requirement the minimum level of protecting antibodies should be there up to six months therefore sponsors have to continue the study and then can claim the actual efficacy of vaccine.

Indigenous development of SARS-CoV-2 killed vaccine by conventional method using Indian isolate in the country is an excellent attempt for controlling the Covid 19 disease. The vaccine so far seems to be good based on the results of controlled clinical trials and its effectiveness will be further come to know over the time after its massive use in vaccination program. It is nice that the said vaccine has been exported to many other countries. Let us hope for its early approval by WHO for emergency use.

Dr.G.R.Soni

On 2021-10-21 14:59:55, user Aditya Bhavar wrote:

Hello,<br /> Where to find exact data of adverse reaction occurred during phase-3 trial as on provided preprint publication data is available in very short and I am specifically interested in studies of adverse reaction occurred during trial phase?

On 2020-09-10 21:35:51, user Nadav Rakocz wrote:

This paper was presented as part of the KDD'20 AI for COVID-19 workshop.<br /> A short talk can be found here: https://insights.kensci.com...

Or here:

https://www.youtube.com/wat...

On 2021-07-11 12:09:52, user Radical Rooster wrote:

The article fails the first test of objectivity: SELF COLLECTED SAMPLES. Scientific research must be rigorous, procedures cannot vary from one person to another. Sampling must be done by only a select group of samplers. In this paper, there were 3,975 samplers.

On 2021-07-17 14:17:15, user killshot wrote:

Unless "efficacy" and "positivity" are more clearly defined, this is meaningless. Eg, using PCR amplification of 35 or greater for diagnosis then using an amplification of 24 for "transmission" would greatly favor vaccine-related prevention of transmission but would be faux data. Also, unless there is randomization for vitamin D levels -- shown to impair virulence if > 36 ng/mL -- the data is also meaningless.

On 2020-09-29 04:18:22, user Melvin Joe wrote:

Its good sign ! ! By wearing the masks we could flatten the curve of affected persons. I am strictly using HY Supplies Inc masks to reduce the infection !!

On 2021-09-01 16:05:27, user 4qmmt wrote:

The phrase "previously infected" is not accurate. According to the study, they were previously positive per PCR test, even though they had info on symptoms.

(2) unvaccinated previously infected individuals, namely MHS members who had a positive SARS-CoV-2 PCR test recorded by February 28, 2021 and who had not been vaccinated by the end of the study period;

Per MoH reports, Israel runs PCR at Ct up to 40. The level of false positives must therefore be taken into account. Since they are unknown, the best estimation, and the only one which makes any sense, is previously positive and symptomatic. Maccabi knows this data. This is even stated in the study:

information about COVID-19-related symptoms was extracted from EMRs, where they were recorded by the primary care physician or a certified nurse who conducted in-person or phone visits with each infected individual.

But the study says

unvaccinated previously infected individuals, namely MHS members who had a positive SARS-CoV-2 PCR test

The fact that they did not take that into account in the study tells you that the numbers of previously PCR + and symptomatic is lower than just positive PCR. In fact, the tell you that of 19 previously PCR+ in the recovered cohort, only 8 were symptomatic.

So, of those 8, Maccabi knows who was symptomatic before, but the paper does not discuss that. Why? Look at the Cleveland Clinic study which found 0 reinfections in their > 1300 previously PCR positive and symptomatic unvaccinated workers.

From page 7 of that study

"The health system never had a requirement for asymptomatic employee test screening. Most of the positive tests, therefore, would have been tests done to evaluate suspicious symptoms."<br /> What this means is that this Maccabi data study is actually setting the lower limit for the multiple of protection of natural immunity over vaccinated, i.e., at least 27X better, and likely orders of magnitude greater than that.

On 2021-09-09 09:59:21, user Patricia wrote:

This is one of a many studies that confirm what we already know, and that the CDC has confirmed, with >50% infected not knowing they are infected and being asymptomatic. Of those with symptoms, 80% are mild with cold-like symptoms and 80% of deaths are elderly. The US has a 1.6% death rate with SARS2, far below Peru's 9.2% or Mexico's 7.8% or Sudan/Syria 's 7.4% - out of 200 countries, the USA ranks #85/86 in the world. That is most likely due to the CDC refusing to support any safe COVID treatments successfully used around the world, the Media and Politicians condemning and smearing safe treatments (even banning use in multiple states), and the DO NOTHING APPROACH of : if you are really sick and need medical attention, quarantine and do nothing until you are on death's doorstep and need an ambulance & ventilator. Physicians that defied the nonsense & political hatred, by saving lives and treating COVID with very safe FDA approved drugs - with the common practice of off label use to treat the symptoms - show extremely low deaths rates. Excluding the CDC's push to use the very expensive Gilead's Remdesivir intravenously in Emergency settings that has a 2/3 success rate and ZERO STUDIES or Vanderbilt's monoclonal antibodies - There are over 1,000 studies with over 35 easily accessible drugs, that ALL PROVE INCREASED IMPROVEMENT. So why did the CDC and Doctors discourage any treatment and push the DO NOTHING UNTIL YOU ARE ALMOST DEAD PROTOCOL? In 21 months, since the first death in January 2020, 12% of the US Citizens have tested positive. >80% have mild to moderate symptoms. We have no idea how many had SARS2 because we have no idea how many more were asymptomatic. However, the CDC estimates 114.7M Americans have had SARS2 and have natural immunity. According to far left Media sites, the worst states to manage COVID are those with >200 cases per million, NY 349, NJ 284, SC 273, NV 237, DE 221, DC 209, MA 207 per million population. The most heavily populated states were the first states to be afflicted with COVID - 39M CA & 30M TX hovered around 97 per million, despite having highly criticized opposite approaches of CA Strict Lockdowns versus Texas Open with common sense guidelines. Since that revelation, COVID tracking was suspended by many or dates cherry picked when it became popular to smear successful states to deflect from the fact that the vaccine do not work and hence - are not a vaccine. They never said it would stop the disease, just somehow promised people wouldn't have as severe symptoms. Like treating flu with medicine to ease the symptoms for recovery. The US dismisses the chaos around the world, MSM refuses to inform us on major protests around the world that have been occurring for months. Why isn't the Oxford Director of Vaccines statement on every paper and news segment? He confirmed the vaccine isn't stopping SARS2 and herd immunity with it is "unachievable" and "mythical". What we do factually know today - is very detailed tracking in Iceland and Israel, who were also very PRO-VACCINE. Vaccinating ~90% of their populations, only to see a far worse spike of cases in SARS2 outbreaks that appeared in July August 2021. The inventor of the new mRNA vaccine publicized problems with using the vaccine, warning of manipulating antibodies that would results in weaker SARS2 variants become more virulent and slipping past the VAX'd immune system, creating a huge rebound of illness. And this summer, we are witnessing his expert analysis come to fruition.

On 2021-12-01 13:15:14, user MWK wrote:

This study is both limited and flawed on many levels. First, it is a retrospective, observational study. That is, no actual serial testing was done. It was a study of past medical records, and that is all.

No regular covid testing was done to see if asymptomatic people in any of the groups were positive. No regular blood testing was done to see if antibodies were present at a higher number than a previous count. A spike in antibody counts from a previous count would indicate a recent covid infection.

These people were not instructed to get tested if they started showing only minor signs and symptoms as they did not even know they were part of a study until after the fact as it was done retrospectively.

So, it is quite possible that people in all three groups did not get tested if they were showing only minor symptoms.

However, it has been argued, and I completely agree, that the least likely group to go and get tested if showing only minor signs/symptoms would be the previously infected group, especially if they had a moderate to serious sickness the first time around.

It was then and still is now a very widely held belief that previous infection protected an individual. Additionally, previously infected people were and are being asked to donate antibodies to help those who were sick, clearly implying protection from repeat infection.

It also has been argued that the vaccinated group would be more likely to get tested if showing only minor signs and symptoms because no one was sure if the vaccine would work against the Delta variant, and vaccinated people were very concerned.

So, simply put, these numbers are not at all accurate, reliable, or valid.

I do not believe this study will ever be published in a peer-reviewed, scientific journal as the scientific community is already finding serious problems with this study,

On 2021-10-15 23:10:46, user Steve wrote:

I agree this study should be final at this point but I'm not sure the peer review process is even a real thing anymore. Maybe I am too cynical but I find it striking that we can spend almost two months debating about why this study wasn't perfect (either way) yet nobody seems able or interested in putting together a better study to address a simple question: How long does natural immunity last? We have tons of stats on who got covid, and what percentage of people are vaxxed but we can't seem to find any of those recovered patients and test them today for antibodies? <br /> The CDC just released their "study" of a few hundred people in Kentucky that they say proves vaccination is better. Hard to take them seriously when a real study involving almost 50,000 patients is dismissed because it isn't perfect.

On 2022-05-02 23:00:41, user Brian Mowrey wrote:

The authors evince no apparent regard for the importance of the interval between PCR+ and serum sample (PDV), especially given the small number of presumed infections among the mRNA-1273-vaccinated in the main analysis, simply remarking

Anti-N seropositivity at the PDV was similar when stratified by median days from illness

Not for the vaccinated, it wasn't (50% vs 32% when stratified, Table 1). Did the authors merely lump both groups together to get around investigating why N-seropositivity was 18% lower in the -5-53 day vaccinated set?

In fact, the same stratification should have been expected to produce different N-positivity rates in the placebo group, had infections been evenly spaced in the -5-53 day interval. Since the authors find only 74% Day 29 N-positivity for placebo participants who are PCR+ on Day 1, and 60% on Day 57 for placebo who are PCR+ on Day 29, it's clear the placebo group isn't defying standard expectations about seroconversion not being instantaneous (bearing in mind a higher false positive rate on Day 1/29 due to screening, obviously) - until the main analysis, when suddenly there is no apparent penalty for near-PDV infections. So maybe there were almost no near-PDV infections in the placebo group (as in, infections skewed toward February due to seasonal patterns) while in the vaccine group the opposite was true (infections skewed to March due to the waning of infection efficacy)?

Thus, both the values for the placebo group and Covid-vaccine group suggest uneven time between PCR+ and PDV. The authors make no comment on this problem and *do not present* a plot of to-PDV-intervals for either group, even though they obviously had full access to that data. This is a glaring oversight at best.

It's not the only one. Have the authors never heard of false positives / base rate fallacy? I doubt it. So why isn't this taken into account, when comparing a group with a frequent outcome to a rare outcome? Among 14.5k participants in both arms, a mere 25 false PCR+ in both groups would be enough to render the main results way off the mark (Placebo: (100% x (1-((.066x648 - 25)/(648-25)) = 97.1% mRNA-1273: 100% x (1-((.593x52 - 25)/(52-25)) = 77.8%). Yet no consideration of the problem is made. The word "false" is not even in the text.

On 2021-04-13 23:43:54, user greenorange041 wrote:

I think it is quite important to differentiate between simple cloth masks and medical / FFP2 masks that were adopted relatively recently and were shown to be more effective in preventing possible contagion. Accounting for this could potentially make a big difference and lead to smaller estimated effects for some measures given that visitors wear more efficient masks.

Another aspect to consider is that some activities banned as a result of the NPIs can be performed both indoors and outdoors. You seem to be aware of this difference, but apparently you are unable to consistently account for it in your analysis. Hence you cannot differentiate between the effect of closing indoor gastronomy and closing only outdoor gastronomy (and hence reopening it) with all necessary hygienic measures in place, which will presumably be far less significant.

The NPIs on your list are also defined quite broadly. In particular you don't consider closing sport facilities and banning even outdoor sports to be a separate NPI. Banning hotel stays for touristic purposes is also not on your list.

Finally, you don't seem to account for season or weather in your analysis. In winter when people tend to have weaker immunity, some measures can indeed make a difference. But in summer the situation can be quite different and some measures could add little no value.

A very sad and disappointing consequence of this study is that it motivates governments to keep in place the same already known plain and undifferentiated measures even though 1) their effect may be largely overestimated in the current situation (by that I first of all mean wide adoption of masks and FFP2 masks in particular as well as limits on the number of visitors / clients that are already in place) and 2) their effect can be different under different conditions (indoor / outdoor activities), however this difference is not analysed.

Another disappointing consequence is that not listing and not analysing certain more detailed measures (closing sport facilities and banning tourist stays in hotels) will most certainly lead to keeping them in place even though the study doesn't provide any explicit evidence in favour of such measures.

To put in in another way: can someone get an answer to the following questions from your study: what is the possible negative (if any) effect of opening retail given that everyone wears an FFP2 mask and there is a limit on the number of clients, what is the possible effect of allowing outdoor sports, of allowing outdoor gastronomy with sufficient hygienic precautions in place (though without express testing), what is the possible effect of opening hotels? And (what is more important), what is the effect of all this given warm temperatures and more daylight? As far as I understand, it is not possible to answer these questions based on your study, but governments may still be tempted to interpret the conclusions as a justification not to relax any measure in a meaningful way.

Of course, correct me if I am wrong.

On 2021-04-29 12:24:41, user Ric wrote:

I think that this estimation strategy is seriously flawed.

The underlying assumption is that Rt is on averge constant over time, unless some measures are taken by the government. This is obviously false, since all epidemic sooner or later ends even without any intervention. Moreover, government actions are obviously taken when the number of cases is already high and Rt could have started to decline on its onw, so you are basically confusing a correlation with causation.

This is seriously concerning. I have already seen articles by general press pushing for more interventions based on this completly unreliable estimations. Please revised your methodology completly or be clear that this is a correlation that does not estimate any effect

On 2021-10-31 15:46:14, user Jan Brauner wrote:

This manuscript has been published now: https://www.nature.com/arti...

On 2020-10-27 16:47:39, user Kamran Kadkhoda wrote:

Low prevalence....low PPV....meaning serology is not reliable.

On 2021-11-12 10:44:25, user Ken wrote:

The next step could be linking the TeKWP to the hospitalization rate in the ICU so to have a real time indicator on the stress that the structure can withstand in relation to the new cases

On 2021-09-08 22:00:39, user Cheeseman wrote:

The authors' statements regarding the effectiveness of universal masking must be in concordance with a systematic review from December 2020 titled "Physical interventions to interrupt or reduce the spread of respiratory viruses" (https://dx.doi.org/10.1002/... "https://dx.doi.org/10.1002/14651858.CD006207.pub5)").

The key discussion element is: "The pooled estimates of effect from RCTs and cluster-RCTs for wearing medical/surgical masks compared to no masks suggests little or no difference in interrupting the spread of ILI (RR 0.99, 95% CI 0.82 to 1.18; low-certainty evidence) or laboratory-confirmed influenza (RR 0.91, 95% CI 0.66 to 1.26; moderate-certainty evidence) in the combined analysis of all populations from the included trials."

This review stands in stark contrast to the authors' position that face masks are effective at mitigating viral transmission. To maintain scientific legitimacy, the authors must decrease the strength of their claims on mask effectiveness in light of this review article.

On 2021-08-05 18:32:18, user Anette Stahel wrote:

Dear moderator,

I've now reviewed, edited and updated my earlier comment to the present study [1]. I hope this will allow for it to be posted.

I'm sorry, but this study is not correct. That is, the pool of people used as denominator when calculating the percentage of COVID-19 infected people who developed CVT and PVT is greatly inadequate. I'll explain what I mean.

In the abstract of the study, it's stated:

"COVID-19 increases the risk of CVT and PVT compared to patients diagnosed with influenza, and to people who have received a COVID-19 mRNA vaccine."

However, when comparing the risk of developing condition X from disease Y with the risk of developing condition X from something else, eg vaccine Z, you first and foremost need to make a correct assessment of how how large the pool of people with disease X is. And to do that, you need to make an estimate. Merely counting the number of people who've sought out primary or secondary care for their symptoms won't do. Not even if you include all the people who were asymptomatic but sought out the care center anyway in order to take a test to see if they were infected (simply because they wanted to know) and then tested positive.

No, you need to include all infected persons in the total pool of people belonging to the health care facility/facilities in question, including the ones who don't go test themselves because of being asymptomatic, or of not having the energy to do it due to their symptoms, or of being into alternative medicine, or of lacking interest/knowledge about the infection et c. There may be many of different reasons. This means you need do make an estimate, otherwise the denominator in the calculation of the percentage who develop condition X from infection Y becomes incorrect.

A study measuring the risk of developing condition X from infection Y using a smaller denominator than one including everyone infected may be useful at times, but it can not be used for comparison with a correctly calculated vaccine risk.

I will use the study Estimation of the Lethality for COVID-19 in Stockholm County published by the Swedish Public Health Agency [2] as an example of a correctly calculated risk, based on an adequately defined denominator. The fact that this is a calculation of the lethality percentage from COVID-19 and not the CVT and PVT percentage is irrelevant, the point is that the same mathematics used in this study should've been applied in the present Oxford University study. From page 13 in the Swedish study, in translation:

"Recruitment was based on a stratified random sample of the population 0-85 years. In the survey we use, the survey for Stockholm County was supplemented with a self-sampling kit to measure ongoing SARS-CoV-2-infection by PCR test. The sampling took place from March 26 until April 2 and 18 of a total of 707 samples were positive. The proportion of the population in Stockholm County which would test positive was thus estimated at 2.5%, with 95% confidence range 1.4-4.2%."

For a complex reason, which I won't go into but is described in detail in the study text, one needs to use a slightly higher percentage when multiplying it with the total number of people in the pool, but that's of minor importance. Anyway, in this study they had to use the figure 3,1169% and when they multiplied it with the number of people in Stockholm County, 2 377 000, they got 74 089. This estimate was then the correct denominator to use when calculating the percentage of people who died from COVID-19 in Stockholm County during this time period.

The numerator was the number of people who died in Stockholm County with a strong suspicion of COVID-19 as a cause, which was 432, no incorrectness there either - as long as a suspected cause number, not a diagnosed cause number, is also used as the numerator when calculating the lethality from the COVID-19 vaccine when the infection lethality and vaccine lethality rates are compared.

So, what they found was that the lethality from COVID-19 in Stockholm County was 0,58%. This is a correct figure, as long as we keep in mind the fact that some of the suspected COVID-19 deaths may later become diagnosed as unrelated to the infection.

The above is thus how the authors of the present Oxford study should've carried out their calculations but they didn't. From their text:

"Design: Retrospective cohort study based on an electronic health records network. Setting: Linked records between primary and secondary care centres within 59 healthcare organisations, primarily in the USA. Participants: All patients with a confirmed diagnosis of COVID-19 between January 20, 2020 and March 25, 2021 were included."

This excludes a considerable amount of infected persons in the total pool of people belonging to all of these primary and secondary care centers, who didn't go test themselves because of a number of reasons (being asymptomatic, being alternative medical, not having the energy or interest for it, et c).

If they'd used the adequate figure in the denominator, the percentage of people established to've developed CVT and PVT from COVID-19 would've gotten vastly lower. However, the percentage of people determined to've developed CVT and PVT from the mRNA COVID-19 vaccines was fully correctly carried out since there are no unregistered vaccinated cases and therefore the registered figure is to be used.

Via the Oxford study's Figure 2 and Table S2 [3], I calculated the following figures: First time CVT cases diagnosed after administration of the mRNA COVID vaccines amounted to 6.6 per million and first time PVT cases after same vaccines amounted to 12.5 per million.

Now, there's a study titled Estimation of US SARS-CoV-2 Infections, Symptomatic Infections, Hospitalizations and Deaths Using Seroprevalence Surveys published by the American Medical Association [4], which has estimated the percentage of infected people in the US looking at roughly the same time period as the Oxford study. From the paper:

"An estimated 14.3% (IQR, 11.6%-18.5%) of the US population were infected by SARS-CoV-2 as of mid-November 2020."

With an infection rate around 14.3%, the estimated number of infected people of the 81 million patients in the healthcare database referred to in the study would've amounted to 11 583 000. This number gives us a hint as for the size of the denominator which should've been used in the calculation instead of the figure of 537 913 confirmed diagnoses.

However, since the Oxford study not only looked at CVT and PVT arising from people having the infection around mid-November 2020 but looked at a much longer time period, from January 20, 2020 to to March 25, 2021, a number far greater than 11 583 000 should be applied. What we need is to estimate how many of the 81 million patients which were infected at least once during these 14 months in question. For the calculation to be really accurate, we need the total, accumulated number of infected people. But since that number isn't found without a very comprehensive and time consuming investigation, we instead have to use the signs ">" ("greater than") and "<" ("less than") here. So, the correct denominator, which should've been used instead of the 537 913 figure, is >11 583 000.

Further, the study says that first time CVT was found in 19 of the patients following COVID-19 diagnosis and first time PVT in 94. This actually means that the rates of CVT and PVT elicited by COVID-19 were much lower than the rates of CVT and PVT elicited by the vaccines. COVID-19 elicited PVT cases, correctly calculated, amounted to <8.1 per million - only about two thirds of the 12.5 per million for the vaccines - and the CVT cases amounted to <1.6 per million - a mere fourth of the vaccines' 6.6.

Interestingly, with their work including this method error, these authors have provided scientific validation of the growing suspicion that the COVID-19 vaccines give rise to thrombocytic complications to a much greater extent than does COVID-19 (which is the opposite of what's stated in the study), because even if the 537 913 figure is inadequate, the other figures in the study are most likely not.

It should also be said that the disclaimer inserted towards the end of the Oxford study by no means can be referred to in order to justify this method error. From the disclaimer:

"However, the study also has several limitations and results should be interpreted with caution. (--) Third, some cases of COVID-19, especially those early in the pandemic, are undiagnosed, and we cannot generalise our risk estimates to this population."

The reason why this passage cannot be referred to, is that 11 000 000 or so omitted cases impossibly can be defined as "some", when the number of denominator cases determined in the study merely constitutes a small fraction (5%) of that figure.

Finally, I'd like to suggest a reading through of the English translation of the Swedish COVID-19 lethality study that I took up in the beginning of my text as a correct, comparative example [5]. This is the main paper that the Swedish equivalent to CDC, the Public Health Agency (Folkhälsomyndigheten), refers to when talking about the COVID-19 lethality here and it's put up on one of the major information pages of their website. I really recommend reading all of it, because it explains so well and in such detail how come this model of denominator calculation without exception must be used in studies like these, which aim to investigate the rate of injuries/complications arising from an infectious illness.

Anette Stahel <br /> MSc in Biomedicine <br /> Sweden

References

1. Taquet, M, Husain, M, Geddes, J R, Luciano, S & Harrison, P J (2021) Cerebral Venous Thrombosis and Portal Vein Thrombosis: A Retrospective Cohort Study of 537,913 COVID-19 Cases medRxiv https://doi.org/10.1101/202...
2. Svenska Folkhälsomyndigheten (2020) Skattning av Letaliteten för Covid-19 i Stockholms Län https://www.folkhalsomyndig...
3. Taquet, M, Husain, M, Geddes, J R, Luciano, S & Harrison, P J (2021) Cerebral Venous Thrombosis and Portal Vein Thrombosis: A Retrospective Cohort Study of 537,913 COVID-19 Cases OSFHome https://osf.io/a9jdq/
4. Angulo FJ, Finelli L, Swerdlow DL. Estimation of US SARS-CoV-2 (2021) Infections, Symptomatic Infections, Hospitalizations and Deaths Using Seroprevalence Surveys (2021) JAMA Netw Open https://jamanetwork.com/jou...
5. The Swedish Public Health Agency (2020) Estimation of the Lethality for COVID-19 in Stockholm County Online translation of [2] https://translate.google.co...

On 2020-04-02 20:13:43, user Sinai Immunol Review Project wrote:

Main findings: The authors analyzed 4000 test results from 28 COVID-19 patients of which 8 were confirmed severe COVID-19 cases and 20 were confirmed cases of mild COVID-19 infection. They found that the overall level of serum CRP increased in all cases irrespective of the disease severity. They observed that serum cystatin C (CysC), creatinine (CREA), and urea, biochemical markers of renal function, were significantly elevated in severe COVID-19 patients compared to mild patients.

Critical Analyses: <br /> 1. Figure duplication in panels G and H of Figure 2 <br /> 2. Survey area is limited to one center.<br /> 3. Small number of participants in the survey.<br /> 4. Elderly people in severe groups and relatively younger people in the milder group. The baseline parameters may differ in both groups, considering the age difference.<br /> 5. Although not clearly stated, this is a cross sectional study and interpretation of results is difficult. The markers that were found to be significantly different between groups are very non-specific. Renal failure and high LDH are not surprising findings in critically ill patients. <br /> 6. There is a very minimal description of the patient's baseline characteristics. It would be important to know for example what were the symptoms at presentation, how long patients had symptoms for before inclusion in the study, duration of hospitalization before inclusion. This would help interpret whether results reflect difference in severity of disease or simply a longer course of disease/hospitalization. <br /> 7. It is unclear what the authors mean in the discussion when they mention “which may be the result of prophylactic use of drug by doctor” (Discussion section, line 6). Type of the drug used is not specified.

Relevance: This study offers insights on some laboratory markers of mild vs severe cases of COVID-19 infection. Glomerular cells highly express ACE2 which is the cellular receptor for SARS-CoV-2, and impaired kidney function might represent a marker of virus-induced end organ damage.

Reviewed by Divya Jha/Francesca Cossarini as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.

On 2020-03-26 02:33:57, user Elisabeth Bik wrote:

Figures 2 and 3 would work better if the letters A, B, C, etc, were replaced by the actual serum marker name. Also, panels G and H in Figure 2 appear to be duplicated (same Y axis label and same graphs). Could the authors check, please?

On 2021-08-09 17:23:34, user vaxpro wrote:

author indicated "The sample size of our trial design meets the minimum safety requirement of 3,000 study participants for the vaccine group, as recommended by the FDA, and WHO guidance". Neither FDA nor WHO recommends sample size for an individual study. sample sizes of the individual studies are driven by the objectives. sample size for approval is a different matter. In fact, what FDA says in the referenced document is that "FDA does not expect to be able to make a favorable benefit-risk determination that would support an EUA without Phase 3 data that include the following, which will help the Agency to assess the safety of the vaccine:...ii. All safety data collected up to the point at which the database is locked to prepare the submission of the EUA request, including a high proportion of enrolled subjects (numbering well over 3,000 vaccine recipients) followed for serious adverse events (SAEs) and adverse events of special interest for at least one month after completion of the full vaccination regimen". also "FDA does not consider availability of a COVID-19 vaccine under EUA, in and of itself, as grounds for stopping blinded follow-up in an ongoing clinical trial". WHO guidance is for general vaccine development with major caveats highlighted in the guidance document. the selective and mis- interpretation of the FDA and WHO guidance is regrettable.<br /> immunogenicity is the primary objective of the study. it's curious why author chose to report full set of IgG data at all visits but NT data only at selected visits.post injection pain was reported in ~20% placebo (saline) subjects, which far exceeded any historic rates in this type of population. although no impact on the interpretation of safety for the active arm within this double blinded trial, the observation should be investigated for potential trial conduct issues and discussed.<br /> author stated "MVC-COV1901 has recently been granted EUA in Taiwan". however this is not a scientific issue for this trial, should not be included in trial report in a peer reviewed journal.

On 2020-04-04 10:53:48, user Statistics wrote:

almost 80% at the cross test(!)... so what about Iris diagnosis? Tuberculosis was over 80 % back in the 50s... so if anyone has time to evalute (complete time table please) I will appreciate; also check the completeness of the waldeyer throat ring of the infected; just for the interest...have thanks and praises

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 2020-04-04 22:05:26, user PhilipandHeidi Kapitaniuk wrote:

Here in France the BCG has been widely used, and we still are losing many people to the Coronavirus. They need to be looking at more than just a few countries. This does not sound serious to me.

On 2020-04-07 16:26:34, user Richard wrote:

they stopped the BCG vaccination in australia in 1982, interesting that the death rates in australia amongst the older people are lower than seen in other countries,

On 2020-03-30 15:52:29, user Rosemary TATE wrote:

Hi, I have just performed a review of this preprint. I hope it is useful. I'm a medical statistician. I'd certainly like to see the next version, and it would be good if you could take my comments on board. I'd be happy to help with the stats if you need it.

On 2020-04-01 18:41:26, user Ronald McCoy wrote:

This totally glosses over China. Huge case numbers with universal BCG vaccination. This is a classic example fo confirmation bias. This article has more holes than a Swiss cheese factory.

On 2020-04-02 17:08:35, user Yaira Ca Ce wrote:

Interesting correlation BUT... Is it strong enough the correlation of data from countries in different stages? Being from Mexico makes me think of the lack of detection in my country. The tests haven't been applied massively, and there are MANY cases of atypical pneumonia and flu. Interesting isn't it? It might be that a low morbility and mortality is due to under developed countries in their first stages or there's not a massive testing. Still interesting to consider it. At the end it will be more realistic to make a comparison.

On 2020-04-05 19:18:00, user Sinai Immunol Review Project wrote:

Main Findings: <br /> The study compares IgM and IgG antibody testing to RT-PCR detection of SARS-CoV-2 infection. 133 patients diagnosed with SARS-CoV-2 in Renmin Hospital (Wuhan University, China) were analyzed. The positive ratio was 78.95% (105/133) in IgM antibody test (SARS-CoV-2 antibody detection kit from YHLO Biotech) and 68.42% (91/133) in RT-PCR (SARS-CoV-2 ORF1ab/N qPCR detection kit). There were no differences in the sensitivity of SARS-CO-V2 diagnosis in patients grouped according to disease severity. For example, IgG responses were detected in 93.18% of moderate cases, 100% of severe cases and 97.3% of critical cases. In sum, positive ratios were higher in antibody testing compared to RT-PCR detection, demonstrating a higher detection sensitivity of IgM-IgG testing for patients hospitalized with COVID-19 symptoms.<br /> Limitations of the study:<br /> This analysis only included one-time point of 133 hospitalized patients, and the time from symptom onset was not described. There was no discussion about specificity of the tests and no healthy controls were included. It would be important to perform similar studies with more patients, including younger age groups and patients with mild symptoms as well as asymptomatic individuals. It is critical to determine how early after infection/symptom onset antibodies can be detected and the duration of this immune response.<br /> Relevance:<br /> The IgM-IgG combined testing is important to improve clinical sensitivity and diagnose COVID-19 patients. The combined antibody test shows higher sensitivity than individual IgM and IgG tests or nucleic acid-based methods, at least in patients hospitalized with symptoms. <br /> Review by Erica Dalla as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.

On 2020-04-05 22:01:17, user Kirsten McEwen wrote:

Can the authors provide power analysis?

On 2020-04-06 11:27:18, user Eleanor Johns wrote:

"Conclusion: The SARS-CoV-2 prevention needs to focus on the screening of asymptomatic patients in the community with a history of contact with the imported population, especially for children and the elderly population." 28% of infected COVID19 individuals are asymptomatic, as PCR and Antibody testing covered a wide population in a Chinese province. We must test EVERYONE.

On 2020-04-06 14:39:20, user Jason Kidde wrote:

It would be interesting to see how blood type applies along age groups with regard to disease severity and death. If the finding is preserved across age groups, this would add muster. Additionally, I'm curious about looking into death rates and severity across geographic regions. For instance how much does blood type explain the death rates in Eastern Europe being that the type A allele is more common in this region, while the type A allele virtually does not exist in South America. Will this result in lower death rates in South America? So far Brazil has a fairly average death rate (4%) compared to other nations whereas Chile's is quite good at 0.7%. I realize that many other factors effect this, principally the testing vs true total disease as well as healthcare infrastructure.

On 2020-03-25 12:29:20, user jenniepoole wrote:

Hi were any of the patients with <br /> type A also checked for RH negative or were they all A + ?? This is important to me.

On 2020-03-25 15:31:18, user Sinai Immunol Review Project wrote:

These authors compared the ABO blood group of 2,173 patients with RT-PCR-confirmed COVID-19 from hospitals in Wuhan and Shenzhen with the ABO blood group distribution in unaffected people in the same cities from previous studies (2015 and 2010 for Wuhan and Shenzhen, respectively). They found that people with blood group A are statistically over-represented in the number of those infected and who succumb to death while those with blood group O are statistically underrepresented with no influence of age or sex.

This study compares patients with COVID-19 to the general population but relies on data published 5 and 10 years ago for the control. The mechanisms that the authors propose may underlie the differences they observed require further study.

Risk stratification based on blood group may be beneficial for patients and also healthcare workers in infection control. Additionally, investigating the mechanism behind these findings could lead to better developing prophylactic and therapeutic targets for COVID-19.

On 2020-03-27 00:50:17, user Simon Brazendale wrote:

I read this paper quickly and feel that people need to keep an open mind either way, the question is 'can this finding be reproduced?'. Certainly meta analysis seems to have been achieved, but the I squared values are greater than 25% for blood groups A and AB, which suggests some heterogenity. Lets see if out this global tragedy other studies can reproduce this as well as the original finding needing further scrutiny. It may be part of a key to better understanding or just a red herring

On 2020-04-03 03:05:56, user Philomena Okeke wrote:

I like more studies and research to be done on this.If group A+ are vulnerable then they should be protected from this Coronavirus.The B and O should really help especially the 0 group. I am sure that more researchers need to provide more evidence on this critical issues. <br /> Thanks

On 2020-04-06 18:50:06, user Theodore Koukouvitis wrote:

Insightful and readily quantifiable. The author is confident enough to make specific, short-term predictions and warn against the danger of a full removal of social distancing measures.

This paper should be peer reviewed and evaluated ASAP.

On 2020-04-06 18:54:14, user Sinai Immunol Review Project wrote:

This study examined antibody responses in the blood of COVID-19 patients during the early SARS CoV2 outbreak in China. Total 535 plasma samples were collected from 173 patients (51.4% female) and were tested for seroconversion rate using ELISA. Authors also compared the sensitivity of RNA and antibody tests over the course of the disease . The key findings are:

• Among 173 patients, the seroconversion rates for total antibody (Ab), IgM and IgG were 93.1% (161/173), 82.7% (143/173) and 64.7% (112/173), respectively.

• The seroconversion sequentially appeared for Ab, IgM and then IgG, with a median time of 11, 12 and 14 days, respectively. Overall, the seroconversion of Ab was significantly quicker than that of IgM (p = 0.012) and IgG (p < 0.001). Comparisons of seroconversion rates between critical and non-critical patients did not reveal any significant differences.

• RNA tests had higher sensitivity in early phase and within 7 days of disease onset than antibody assays (66.7% Vs 38.3% respectively).

• The sensitivity of the Ab assays was higher 8 days after disease onset, reached 90% at day 13 and 100% at later time points (15-39 days). In contrast, RNA was only detectable in 45.5% of samples at days 15-39.

• In patients with undetectable RNA in nasal samples collected during day 1-3, day 4-7, day 8-14 and day 15-39 since disease onset, 28.6% (2/7), 53.6% (15/28), 98.2% (56/57) and 100% (30/30) had detectable total Ab titers respectively Combining RNA and antibody tests significantly raised the sensitivity for detecting COVID-19 patients in different stages of the disease (p < 0.001).

• There was a strong positive correlation between clinical severity and antibody titer 2-weeks after illness onset.

• Dynamic profiling of viral RNA and antibodies in representative COVID-19 patients (n=9) since onset of disease revealed that antibodies may not be sufficient to clear the virus. It should be noted that increases in of antibody titers were not always accompanied by RNA clearance.

Limitations: Because different types of ELISA assays were used for determining antibody concentrations at different time points after disease onset, sequential seroconversion of total Ab, IgM and IgG may not represent actual temporal differences but rather differences in the affinities of the assays used. Also, due to the lack of blood samples collected from patients in the later stage of illness, how long the antibodies could last remain unknown. For investigative dynamics of antibodies, more samples were required.

Relevance: Total and IgG antibody titers could be used to understand the epidemiology of SARS CoV-2 infection and to assist in determining the level of humoral immune response in patients.

The findings provide strong clinical evidence for routine serological and RNA testing in the diagnosis and clinical management of COVID-19 patients. The understanding of antibody responses and their half-life during and after SARS CoV2 infection is important and warrants further investigation

On 2020-04-07 23:26:41, user Sam Raredon wrote:

Here is a video briefly showing and describing the technique:

PReVentS Circuit Demonstration - COVID-19 / Yale / Niklason Lab<br /> https://www.youtube.com/wat...

On 2020-04-07 23:34:20, user Karl Riley wrote:

It's worth looking at the benefits of a strategic infection variant, whereby those known to be at least risk of death are exposed to the virus in a controlled environment and then 'released' back into the general population thereby facilitating herd immunity. You could even have those, then known to be immune, as a form of shield in areas looking imminently vulnerable (hospital admission figures could be part of the data used for this) to the spread of the infection.

On 2020-04-08 10:19:35, user Rosemary TATE wrote:

Hello, thank your for this interesting article.<br /> Could you please upload the relevant checklist. I believe this is PRISMA? I cant see this anywhere. You will need this if you are intending to publish.

On 2020-04-11 08:07:44, user Jeff Aronson wrote:

This looks like an interesting study, with an important warning about the combination of hydroxychloroquine + azithromycin, which people are beginning to use. But why misleadingly title the paper "Safety of hydroxychloroquine …" when what is being reported is serious adverse events, i.e. unsafety? I hope that when the authors prepare their paper for peer review, they will use a more accurate description.

On 2020-04-13 12:22:24, user Dr. Phillips wrote:

1. In the methods for the ELISA it says that "Serial dilutions of serum and antibody samples were prepared in ..." What was the antibody that was serially diluted?
2. In the figures of AUC, what do the horizontal bars represent? For some it looks like a mean but others, e.g. IgM in Fig 3A, it doesn't.

On 2020-04-15 14:24:47, user Greg Potter wrote:

Any data that looks at HCQ efficacy in COVID-19 patients before they reach the severity of pneumonia?

On 2020-03-14 19:08:41, user Marcia Walker wrote:

This is so interesting, thank you for this! My question is - why did you start with January and not December? The first known case was traced to 1 December, there may have even been cases before then, surely it is possible that it already started spreading internationally before the end of December?

On 2020-03-29 21:18:16, user Randy k wrote:

This study is guessing at how many people were infected. The whole formula is based on a guess.

On 2020-03-20 22:45:31, user Sinai Immunol Review Project wrote:

The aim of this study was to identify diagnostic or prognostic criteria which could identify patients with COVID-19 and predict patients who would go on to develop severe respiratory disease. The authors use EMR data from individuals taking a COVID-19 test at Zhejiang hospital, China in late January/Early February. A large number of clinical parameters were different between individuals with COVID-19 and also between ‘severe’ and ‘non-severe’ infections and the authors combine these into a multivariate linear model to derive a weighted score, presumably intended for clinical use.

The paper is lacking some crucial information, making it impossible to determine the importance or relevance of the findings. Most importantly, the timings of the clinical measurements are not described relative to the disease course, so it is unclear if the differences between ‘severe’ and ‘non-severe’ infections are occurring before progression to severe disease (which would make them useful prognostic markers), or after (which would not).

This paper among many retrospective studies coming from hospitals around the world treating individuals with COVID-19. In this case, largely because of the sparse description of the study design, this paper offers little new information. However, studies like this could be very valuable and we would strongly encourage the authors to revise this manuscript to include more information about the timeline of clinical measurements in relation to disease onset and more details of patient outcomes.

This review was undertaken as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.

On 2020-03-21 19:23:25, user KnowItAll wrote:

For figure 1c, it would be useful to include the number of genomes sampled from each country. The figure makes it seem like there are large differences in the distribution of viruses between countries, but there is only 1 sequence from Sweden, 5 from Italy, 9 from south Korea, vs 25 from the US.

On 2020-03-27 13:07:45, user Guido Marco Cicchini wrote:

Very interesting. However my gut feeling is that a rock solid <br /> analysis will only be possible once the full history of contagion within<br /> the ship is tracked down. For instance people of the crew who work on <br /> the maintenance of the mains ervices of the boat (such as cooking and <br /> cruising) share little space with the people crusing and relaxing. As it<br /> is likely that these workers are young, this cast a whole new <br /> interpretation on the number of contagion within the younger ages <br /> ranges.

Second point is that the number of deaths (fortunately) <br /> has been quite low. This poses a bit of an issue if one wants to <br /> extrapolate from this data. One viable option which has been proposed by<br /> several people in these days it has been to rely on the data of people <br /> with severe symptoms who needed ICU. Typically they are higher and thus <br /> enable more solid conclusions. IMHO the paper could be more solid if <br /> also this metric were included. <br /> Lastly, out of curiosity it <br /> would be interesting to compare the number of fatalities in Diamond <br /> Princess in February to those of other cruising boats (say during Feb <br /> 2019) across the world. I assume that these are quite lower (it is <br /> unlikely that there are more than 5 deaths across 3000 people in a few <br /> weeks of time).. yet it could be interesting. If stats for one months <br /> are too low and unreliable one may want to enlarge the sampling period <br /> to six-months

On 2020-04-01 15:48:07, user mendel wrote:

According to wikipedia, 12 passengers have died by now, 1 in her 60s, 6 in their 70s, 3 above 80, and 2 with unreported age. Assuming the unreported ages were in the 60s and 70s age group, we'd have a distribution of 2-7-3 deaths for the top age groups; Table 2 in this paper expects 2.7-7.6-4.3 for these groups based on naive case fatality rates from the Chinese data. That's fairly close, invalidating the paper's claim that the Chinese data must be off by a large margin.

Given the small sample sizes of the cruise ship, the observed deaths are not at odds with the assumption that the Chinese nCFRs hold: if you establish CFRs for these 3 age groups and error bars for those, then the observed data supports the Chinese findings within the 50% confidence error bars, and the whole ship mortality as well.

On 2020-03-27 15:14:04, user Kevin Hall wrote:

Why does this model calculate deaths as being proportional to the time-delayed fraction of the population not susceptible to infection z rather than the time delayed fraction of the infected population y? See equation 3.

On 2020-03-27 21:12:30, user V. Cheianov, Esq. wrote:

Dear Authors,

your model contains a parameter psi, which is the mean time from infection to death.The way you include this parameter in your calculations (using the retarded value of z times rho times theta) implies that the number of deaths is exponentially sensitive to fluctuations in psi. In order to properly take into account such fluctuation within the population, you need to average the exponentially increasing z(t-\psi) over the probability distribution of psi, P(psi)

In your table 1 you claim that according to Ref [14] psi obeys Gaussian normal distribution <br /> with with M= 7 days and SD =2 days.

In fact, Ref [14] gives the distribution function of the days from onset of illness to death, <br /> which is a log-normal distribution Fig 1 of Ref[14] <br /> with lognormal mean 14.5 and lognormal SD 6.7 without right-truncation (Table 1 of Ref[14]) and lognormal mean of 20.2 days and lognormal standard deviation of 11.6 days <br /> for the right-truncated fit (Table 2 of Ref[14]). <br /> This distribution has to be further amended by the <br /> incubation period and its distribution (with a much smaller SD)

None of these values/distributions look remotely similar to the Gaussian <br /> normal distribution with M=7 and SD=2. Would you please explain how you <br /> arrived at the values and the distribution given in your Table 2.

Thank you very much.

On 2020-03-29 10:45:38, user Bob O'Hara wrote:

Someof us had some problems with this manuscript, so we wrote a response: https://doi.org/10.32942/os...

Abstract: The ongoing pandemic of the severe acute respiratory syndrome <br /> coronavirus 2 (SARS-CoV-2) is causing significant damage to public <br /> health and economic livelihoods, and is putting significant strains on <br /> healthcare services globally. This unfolding emergency has prompted the<br /> preparation and dissemination of the article “Spread of SARS-CoV-2 <br /> Coronavirus likely to be constrained by climate” by Araújo and Naimi <br /> (2020). The authors present the results of an ensemble forecast made <br /> from a suite of species distribution models (SDMs), where they attempt <br /> to predict the suitability of the climate for the spread of SARS-CoV-2 <br /> over the coming months. They argue that climate is likely to be a <br /> primary regulator for the spread of the infection and that people in <br /> warm-temperate and cold climates are more vulnerable than those in <br /> tropical and arid climates. A central finding of their study is that <br /> the possibility of a synchronous global pandemic of SARS-CoV-2 is <br /> unlikely. Whilst we understand that the motivations behind producing <br /> such work are grounded in trying to be helpful, we demonstrate here that<br /> there are clear conceptual and methodological deficiencies with their <br /> study that render their results and conclusions invalid.

On 2020-03-29 23:51:59, user Shuang Gao wrote:

Just like to point out that <br /> "the latest estimates of the death risk in Wuhan could be as high as 20% in the epicenter of the epidemic whereas we estimate it ~1% in the relatively mildly-affected areas“<br /> This epicenter here means the epcienter in Wuhan not Wuhan itself. This expression is different from saying <br /> "the latest estimates of the death risk in Wuhan, the epicenter of the epidemic, could be as high as 20% whereas we estimate it ~1% in the relatively mildly-affected areas"<br /> Wuhan is big and impact of Covid 19 is different in different parts of Wuhan.

On 2020-03-30 16:54:02, user Anton Surda wrote:

How can I get the model available online.

On 2020-03-30 21:04:58, user Sinai Immunol Review Project wrote:

Keywords<br /> death biomarkers, cardiac damage, Troponin, Blood type, respiratory failure, hypertension

Summary<br /> This is a retrospective study involving 101 death cases with COVID-19 in Wuhan Jinyintan Hospital. The aim was to describe clinical, epidemiological and laboratory features of fatal cases in order to identify the possible primary mortality causes related to COVID-19.

Among 101 death cases, 56.44% were confirmed by RT-PCR and 43.6% by clinical diagnostics. Males dominated the number of deaths and the average age was 65.46 years. All patients died of respiratory failure and multiple organs failure, except one (acute coronary syndrome). The predominant comorbidities were hypertension (42.57%) and diabetes (22.77%). 25.74% of the patients presented more than two underlying diseases. 82% of patients presented myocardial enzymes abnormalities at admission and further increase in myocardial damage indicators with disease progression: patients with elevated Troponin I progressed faster to death. Alterations in coagulation were also detected. Indicators of liver and kidney damage increased 48 hours before death. The authors studied the deceased patients’ blood type and presented the following results: type A (44.44%), type B (29.29%), type AB (8.08%) and type O (18.19%), which is inconsistent with the distribution in Han population in Wuhan.

Clinical analysis showed that the most common symptom was fever (91.9%), followed by cough and dyspnea. The medium time from onset of symptoms to acute respiratory distress syndrome (ARDS) development was 12 days. Unlike SARS, only 2 patients with COVID-19 had diarrhea. 98% presented abnormal lung imaging at admission and most had double-lung abnormalities. Related to the laboratorial findings some inflammatory indicators gradually increased during the disease progression, such as IL-6 secretion in the circulation, procalcitonin (PCT) and C-reactive protein (CRP), while platelets numbers decreased. The authors also reported an initial lymphopenia that was followed by an increase in the lymphocytes numbers. Neutrophil count increased with disease progression.

The patients received different treatments such as antiviral drugs (60.40%), glucocorticoids, thymosin and immunoglobulins. All patients received antibiotic treatment and some received antifungal drugs. All patients received oxygen therapy (invasive or non-invasive ones).

Limitations<br /> This study involves just fatal patients, lacking comparisons with other groups of patients e.g. patients that recovered from COVID-19. The authors didn’t discuss the different approaches used for treatments and how these may affect the several parameters measured. The possible relationship between the increase of inflammatory indicators and morbidities of COVID-19 are not discussed.

Relevance<br /> This study has the largest cohort of fatal cases reported so far. The authors show that COVID-19 causes fatal respiratory distress syndrome and multiple organ failure. This study highlights prevalent myocardial damage and indicates that cardiac function of COVID-19 patients should be carefully monitored. The data suggest that Troponin I should be further investigated as an early indicator of patients with high risk of accelerated health deterioration. Secondary bacterial and fungal infections were frequent in critically ill patients and these need to be carefully monitored in severe COVID-19 patients. Differences in blood type distribution were observed, suggesting that type A is detrimental while type O is protective – but further studies are needed to confirm these findings and elucidate if blood type influences infection or disease severity. Several inflammatory indicators (neutrophils, PCT, CRP and IL-6, D-dimer) increased according to disease severity and should be assessed as biomarkers and to better understand the biology of progression to severe disease.<br /> Reviewed as part of a project by students, postdoctoral fellows and faculty at the Immunology Institute of the Icahn School of Medicine at Mount Sinai

On 2020-03-31 15:11:10, user Lawrence Mayer wrote:

A study claiming to be a clinical trial but is not peer reviewed and will not release data is not worth posting. China now claims to have 24 clinical trials suppporting the use of HCQ, Not one of them is published with details and no data has been released. Please ignore this claim. The College of Clinical Toxicologists issued a warning yesterday about HCQ for Covid19

On 2020-03-31 16:18:25, user Nicholas DeVito wrote:

The authors have provided the incorrect Trial ID in their abstract (ChiCTR2000030679), however it appears that the correct Trial ID is provided in the full text (ChiCTR2000030697).

In addition, the protocol link provided (http://www.chictr.org.cn/ed... "http://www.chictr.org.cn/edit.aspx?pid=50781&htm=4)") is not the publicly accessible version of the registry protocol entry and should be replaced with (http://www.chictr.org.cn/sh... "http://www.chictr.org.cn/showprojen.aspx?proj=50781)").

Ensuring correct record linkage is important as evidence is gathered and disseminated on the COVID-19 pandemic.

On 2020-04-01 15:47:26, user JR Davis wrote:

Table 3 and 4 and 5 are all missing. Text mentions non-CoVID respiratory pathogens (n=10) also tested for, and listed in "Table 3"....with the additional Primer list in Table 4.<br /> However, both Table 3, 4, and 5 NOT provided in the PDF....only Table 1 and 2 found at the end of the document.<br /> Can you provide missing tables 3,4,5?

On 2020-04-01 16:49:01, user Joe Ledbetter wrote:

Are these projected excess deaths? In other words, are the expected baseline deaths for each age group subtracted from the projected covid 19 deaths?

On 2020-04-01 20:41:26, user Karen wrote:

Ferguson no longer assumes an IFR of 0,9%. Their latest paper (10.1101/2020.03.09.20033357) calculates an IFR of 0,66% for China.

On 2020-04-01 21:21:41, user anand maurya wrote:

I think it would be good to associate this research with other available data, like difference between mobile user subscription, reduction in network traffic. This might atleast give some additional data point to understand what or where have those users disappeared.

On 2020-04-22 23:23:26, user Eric Solrain wrote:

"It was also reported that the maximum outdoor air supply was operated during the quarantine<br /> period." Is this 100% fresh air with no return? The referenced article (https://www.jstage.jst.go.j... ) notes that 100% fresh air is the norm, but for energy efficiency cabins are reduced to 30%. Full economizer mode (at 100% fresh air) is also a common energy saving measure.

On 2020-06-22 23:20:07, user Charles Warden wrote:

Hi,

Thank you very much for putting together this pre-print and database for Polygenic Risk Scores.

I took a quick look at the website, but it is possible that I might have overlooked something:

Is there currently a way to apply these scores to your own samples (and see the distribution of scores from other samples that have been tested)? If not, is this something that you plan to add in the future?

I have done some testing with PRS percentiles, but I wasn't very impressed with what I have tested so far:

http://cdwscience.blogspot....

So, I was curious how these other scores might compare.

Thank You,<br /> Charles

On 2020-06-27 20:40:02, user many wrote:

Major comments:<br /> The paper’s primary claim is not directly supported by the data shown in the manuscript, due to insufficient statistical analyses. The authors can improve their analyses to support their claim. Describing them below.

Figure 2 is key to supporting the primary claim of this manuscript. As of now, Figure 2a only shows a bar graph for each data point. I would recommend using a box plot that can represent the median, standard deviation, 25, 75 percentile values, etc.

The key sentence that brings out the claim (page 7, last line), uses a Ct> 26. Could you provide a reason for using the cut-off to be 26?

Along with the previous comment, when does the Ct value reach 26 for mild and severe patients? This question can be answered by redoing figure 2b. Currently, the figure shows scattered data points roughly 10. But, as I understand from figure 1a, there is possibly more data than what is represented in figure 2b. Therefore, I again recommend using a box plot in figure 2b to represent the true statistical variation of Ct over time.

To support the claim that symptom severity is more important than Ct or time since symptom onset, the CPE should be higher (with a p<0.01) in severe symptom patients than mild symptom patients, irrespective of the Ct or time since onset. The latter (CPE vs time since symptom onset) needs to be plotted in a box plot for better understanding.

Minor points:

Provide p-value on the graphs.

Use of --% “versus” --% sentence structure is misleading. For instance, in the results section, the last sentence, “… outpatients and hospitalized… are: 47% versus 18%...” Is 47% associated with outpatients? In which case, you’d be contradicting your own claim.

On 2020-06-29 16:23:37, user David Eyre wrote:

Until an updated version is posted by medRxiv - you can find the version with the figures displayed correctly here - https://unioxfordnexus-my.s...

On 2020-07-02 20:44:39, user Joel Silveira wrote:

All of this may be due to 5-HT (serotonin) resulting from platelet degranulation, including NET.

On 2020-07-14 06:27:42, user jaswinder singh maras wrote:

Please read and suggest

On 2020-04-17 18:38:09, user Marm Kilpatrick wrote:

Can you please provide a more detailed breakdown of the ages of those sampled and the general pop? Grouping 19-64 year-olds obscures a potentially enormous amount of variation. It's also not clear why you didn't adjust estimates for age. The justification appears to be that your sample sizes were too small. Without adjustments for age it's not clear how one can make an accurate estimate of fatality ratios given the substantial age effects for COVID-19.<br /> Can you also present the results by age group (with finer age groupings - e.g. decades or 5yr incements)?<br /> Could you also present results based on prior symptoms? It seems quite likely that individuals w/ COVID-19 symptoms would be more likely to be recruited into study.<br /> Could you report the sensitivity results for your known samples at Stanford by IgG and IgM like you present the manufacturer data? This would help the reader understand the discrepancies.

Finally, it seems likely that socio-economic status of Facebook users and non-facebook users likely differs. It doesn't appear that you collected this data and yet it seems like it could significantly influence the results. Can you discuss this issue in Discussion?

Thanks!<br /> marm

On 2020-04-17 19:58:00, user John Ryan wrote:

50 out of 3,300 study participants tested positive for antibodies. This is actually a very low number given that lead researcher Dr. Eran Bendavid has been floating the notion that herd immunity has already been achieved in California, which is why the mortality rate is so low. Dr. Bendavid wrote an opinion piece in the WSJ on March 24 arguing that the case prevalence is much higher than revealed by testing and based on that analysis, the U.S. would see a maximum of between 20K & 40K deaths. We will pass that upper level this weekend.

This study did not have a random sample but a convenience sample drawn from Facebook users, many of whom believed they had already had CV-19. Lots more research to be done before any sweeping conclusions are drawn.

On 2020-04-18 02:41:54, user Andy wrote:

Based just on the 50x number, many places in New York already have herd immunity. Everyone in Westchester (currently 2,253 cases per 100,000 people) should already have it.

https://www.nytimes.com/int...

Based on the 85x number, more than everyone in New York City (currently 1,458 cases per 100,000 people) also already have it.

As noted in the paper, "bias favoring those with prior COVID-like illnesses seeking antibody confirmation [is] possible."

The bias has to be very very significant, if you think about why anyone would venture out to risk exposure to be tested during the state's Stay At Home order. In other words, if you do not believe you have been exposed previously, why would you even go -- I wouldn't.

On 2020-04-18 14:59:28, user Julie Larsen Wyss wrote:

I was one of the 3300 that was tested. At this time I am told that those that tested positive for the antibody have not yet been informed. Any ideas why they have not informed the 50 or so positive participants yet even though they have released the study to the public?

On 2020-04-19 00:58:33, user dixon pinfold wrote:

If you read between the lines of the final paragraph of the Discussion, you can perhaps guess at some of the motivation behind the study.

No other antibody seroprevalence studies had been started in the US prior to this one, and the study's authors may have thought that it was high time, somewhat-dubious antibody tests and barriers to random sampling be damned.

On rare occasions, owing to a sense of urgency, the farmer may think it necessary to plant a crop despite the ground not really being prepared.

If others are white-hot with indignation at the very idea, it's their right, and from their side of it they are probably correct. For my part, I'm less than sure.

On 2020-04-19 05:23:12, user John Dixon wrote:

This may be stupid, but if the ad specificies what the test is for, then doesn't that render it immediately unrepresentative? If it says it's for Covid-19, then won't people be more likely to go who have had cold or flu symptoms recently and are worried they may have had it? And so the sample pool would tend to have more positives than a purely random selection of the population. Therefore the study would underestimate the fatality rate. Am I missing something? To get a random sample, wouldn't you have to leave out any specifics of what the test is for?

On 2020-04-19 18:41:29, user Dean Karlen wrote:

The authors are reporting incorrect confidence intervals because they to not correctly treat the unknown false positive rate. Use the manufacturer data for false positives (2 out of 371 known negatives) to give the posterior probability for the false positive rate (fpr) which is proportional to Binomial(2, 371, fpr). With this, calculate the 95% CL interval using the exact approach (Neyman). The correct interval, for the unadjusted case is:

[0.00% - 1.53%]

The authors report an incorrect interval for this case: [1.11% - 1.97%].

Because the unadjusted case is such a simple problem to interpret, there is only one correct treatment to produce the 95% central confidence interval. Done correctly, and reporting the correct intervals, this paper would not gain any attention at all. Please ignore this paper. It is only getting attention because the authors made serious mistakes in the analysis. The authors should retract this erroneous paper.

Python code with this calculation will be provided to anyone on request. I have contacted the lead author, pointing out their error in statistical analysis. I have received no response.

On 2020-04-20 07:07:53, user clever trevor wrote:

The Achilles heel of this study is the specificity of the serology test.

On the manufacturer's own data, they tested 371 blood samples stored from the pre-COVID era, and got 2 false-positives.

False positives are real problem on population testing. if on that crude data they over-estimate the prevalence of sero-positivity by 0.66%points, that throws the whole calculation into doubt.

the authors did their own testing for specificity, but on only 30 samples, and, inexplicably, those 30 samples were from hip-surgery patients. Hip surgery patients tend to be old, *and* therefore they tend to have generally lower circulating levels of immunoglobulin,

https://www.ncbi.nlm.nih.go...

so a cohort of hip-surgery patients is *the wrong group* to look at if you want to stress-test the specificity of your assay.

This study needs to be repeated with much stronger specificity evidence in the assay.

On 2020-04-21 06:51:43, user Vladimir Lipets wrote:

Well, from statistical/math perspective there are significant errors in results interpretation.Based on calibration experiment (2 FP of 271), authors assumed that FP range is very low. However, it is incorrect, obviously. And I’m not the first one who point out about this mistake.

Since calculating posterior probabilities combining Binomial distribution seems to be little bit tricky, I spend 15 minutes and did Monte-Carlo experiment as follows: A) Randomly selected FP probability in range 0-1% B) Simulated 270 experiments with FP probability chosen C) If exactly 2 FP results were obtained, then the main test of 3300 iteration was simulated. Steps A,B,C where repeated 1M times, to get the results. (well be glad, if somebody corrects me, if there are mistakes in this approach)

Finally, I got FP distribution which estimates probability of having more than 50 FP in 3300 (random?) candidates is about 20%. Too high... Having more than 40 is 33%

It is very confusing that these results are wrong, considering the importance of these results to the… well, whole world!

On the other hand, significant infection rate, still remains maximum likelihood.

Moreover, for me, hypotheses of higher infection rate, still seems very reasonable, let’s wait for more studies to come. As far as I understood, author want to repeat this experiment in NY

P.S. I think,I will wait for these results even more then for last episode of GoT. I hope it will not be disappointing, like this one ))

On 2020-04-22 00:40:03, user Unko J wrote:

It's nice to read below what essentially IS the 'peer-review' for this pre-print online paper! I wish I had read these comments last night before having a heated debate with my fellow quarantinees. My point was how could these possibly be 2%-4% of the population that is positive and yet Santa Clara has only 83 deaths? These divergent sets of data can't really exist in one universe, unless either we're wildly wrong about either a) the mortality rate or b) how many people can be asymptomatic and test positive with an Ab test. So yeah, between cross-reactivity against non-Covid antibodies and other false positives, I think we've decided to reject this paper. And aren't some of the authors the same on both papers?

On 2019-07-20 05:46:57, user Guyguy wrote:

EVOLUTION OF THE EBOLA EPIDEMIC IN THE PROVINCES OF NORTH KIVU AND ITURI

Friday, July 19th, 2019

The epidemiological situation of the Ebola Virus Disease dated 18 July 2019:<br /> Since the beginning of the epidemic, the cumulative number of cases is 2,546, of which 2,452 confirmed and 94 probable. In total, there were 1,715 deaths (1,621 confirmed and 94 probable) and 721 people healed.<br /> 478 suspected cases under investigation;<br /> 14 new confirmed cases, including 6 in Beni, 5 in Mandima, 1 in Katwa, 1 in Mabalako and 1 in Mambasa;<br /> 10 new confirmed cases deaths:<br /> 6 community deaths, 2 in Beni, 2 in Mandima, 1 in Mabalako and 1 in Mambasa;<br /> 4 CTE deaths, 2 in Butembo, 1 in Katwa and 1 in Mabalako;<br /> 3 people healed out of Beni ETC

.167 152 Vaccinated persons

76,319,878 Controlled people<br /> 80 entry points (PoE) and operational health checkpoints (PoC).

138 Contaminated health workers<br /> One health worker, vaccinated, is one of the new confirmed cases of Mandima.<br /> The cumulative number of confirmed / probable cases among health workers is 138 (5% of all confirmed / probable cases) including 41 deaths.

Source: Ministry of Health press team on the state of the response to the Ebola epidemic in the Democratic Republic of the Congo

On 2020-04-18 06:11:39, user Sergey Morozov wrote:

The manuscript provides the readers with the results of retrospective analysis of different regimens of treatment of SARS-Co-V2 infected patients in a single centre in Wuhan, China.<br /> Despite several limitations, properly discussed by the authors, the described results are very actual and may impact clinical practice as COVID-19 pandemic has not yet reached its peak in most of countries, no universal and highly effective treatment was found, whereas some of the proposed remedies showed their efficacy in-vitro only. The study is methodologically correct. However, if possible, I would suggest to add the information on whether selection criteria for study population were applied (all patients admitted to the hospital and who received interferon (IFN), IFN+ Umifenovir (ARB), or ARB treatment, or only some of them).<br /> The authors convincingly proved that inhaled IFN-?2b affect 2 major ways of pathogenesis, namely, viral replication and host's immune response (IL-6), while effects of ARB remain questionable.<br /> A pilot nature of the study requires confirmation of the results in randomized controlled multicentre trials with greater number of patients enrolled. Still, at the present state it may let to avoid waste of the financial sources to the treatment regimens that seem to have poor clinical effect. <br /> Minor remark: please, consider avoidance of the use of the trade name of investigated product (arbidol), if possible. The paper is very well-organized, every statement is logical, weighed and supported with objective grounds.<br /> COI statement: I have no conflict of interest in the regard to this review.

On 2020-04-18 15:24:11, user Robert Clark wrote:

This is potentially a bombshell report, of especial importance for health care workers, showing 100% protectiveness against COVID-19 using interferon. A flaw in the report though is that while it gives a total number of health care workers who didn't take the drug contracting COVID-19, it doesn't compare that to the total of all health care workers. So we can't make the comparison in percentage terms of how many on interferon who contracted the disease (0% according to this report) compared to those not on interferon who contracted it.

Robert Clark

On 2020-04-19 05:52:47, user AlanCarrOnline wrote:

So a single 14 day lockdown is not enough. How many went on to develop symptoms in the next 14 days?

On 2020-04-19 15:29:27, user Tom Grys, PhD wrote:

Unless I'm missing something, there are concerns about the methods used to infer Viral Load. A linear regression is NOT appropriate for Ct. Each 3.32 cycles is 10x more. They need a log regression, or assign a dummy value to their LOD and calculate using relative numbers. Maybe the authors can clarify the methods to help us better understand the data? I am willing to believe the conclusions (Biology never fails to surprise us), but the data must be shown a different way to make it more clear whether VL correlates with anything.

On 2020-04-19 09:51:55, user Arne Elofsson wrote:

Just a note (from Arne): It is clear that our predictions for Sweden (done a week ago) were quite wrong (as it earlier models), the exponential increase in deaths in Sweden has not materialized. This will be made clear in a revised version - along with further analysis.

On 2020-04-19 12:52:50, user David Steadson wrote:

The model uses a base of 200 cumulative deaths for March 31 to calibrate. FHM data as of today (April 19) reports 329 cumulative deaths for that date, a figure 64.5% higher - and that data is still subject to change, with 2 deaths being added as recently as yesterday. The doubling time used is also inaccurate based on more up to date date, though not as significantly.

Recalibration would appear necessary.

On 2020-04-20 10:54:46, user hjqq819 wrote:

It seems to conflict with this study: https://smartairfilters.com...

On 2020-10-26 17:59:08, user Meng-Ju Wu wrote:

Hi! It is interesting to read the paper in discussion for EVs to differentiate ALS from healthy and diseased groups. And I want to share my thought on the study.

I think the main contribution of the study includes the purification of EVs with the nickel-based isolation compared to the conventional methods that makes the analysis of specific EV parameters highly sensitive and reliable. If the EVs are reliably differentiate ALS patients from healthy and diseased group, clinical assessment with the blood test will significantly shorten the diagnosis time for ALS and that the treatment may be started as early as possible. In addition, if biomarkers are available to detect ALS patients, it means that we can develop the treatment specific to ALS using their unique properties. Patients can avoid costly and lengthy process of ALS diagnosis.

I have two questions considering the methods. First, why was the supernatant from human plasma diluted in filtered PBS once but the serum from mice required 10 times for dilution? Second, what was the temperature and humidity condition for the incubation of activated charged agarose beads in NBI? I think the time to use the obtained serum would be the limitation of this approach. The content of the EVs might be changed if the centrifuged plasma samples are not immediately used. Such compositional change may be subject to the storage condition and the degradation rate of each specific proteins. It may also vary among species. Therefore, a specific time period to analyze the plasma should be strictly regulated.

In general, I think there are no major grammatic or spelling errors. However, the content may be modified in order to make it more logical and convincing to read. In the introduction part, I think it is important to summarize how is ALS diagnosed clinically. If the readers are informed that electrophysiologic diagnosis takes longer time and effort and make the diagnosis, they would appreciate the value of blood test to detect suspected ALS patient in prodromal state. In the last paragraph of the introduction, it is not reasonable to mention that the study results suggesting EVs are food biomarkers. It should be mention in the discussion or conclusion section. In the material section, the time of patient inclusion was missing. In the animal model, the paper should mention why only female mice with SOD1G93A and male mice with TDP-43Q331K were studied. Also, the timing to study the two different genes as well as the number of the mice were concerning to interpret the results. I want to suggest making a visual diagram on the machine learning technique. You did a great job in comparing the difference between ultracentrifugation and NBI using EV-like liposomes. As such, I want to suggest applying the same comparison onto the animal model to test the reliability of the using the NBI method alone in the paper. The results and the discussion are well-written and consistent with the tables and figures provided

On 2020-04-21 21:15:32, user Iyad Sultan wrote:

Patients who are sicker are more likely to get HQ or HQ+AZ and are more likely to die. Those who got the combination were 50% likely to get mech vent. The only message is that combination is superior and NO HQ alone. Otherwise, this is a biased study that misses the point - sorry!

On 2020-04-22 00:35:27, user Eric H wrote:

The Hazard Ratio confidence intervals in Table 5 of the report shows that the findings of this study are not significant. That plus the uncertainties in the Propensity Score Matching method make it even worse. I noticed the HCQ group contained a substantially higher proportion of high blood pressure and diabetic w/complications than the control group. Worst of all, they apparently did not interview even one doctor to ascertain the range of Tx criteria used.

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 2020-04-22 03:57:44, user IanM wrote:

Hi,<br /> Could you explain how you performed your quantitative RT-PCR?<br /> Also, could you comment on whether a recombinant or plaque purified version of each virus carrying a mutation of interest may increase the strength of these in vitro observations? Cheers!

On 2021-01-23 14:32:57, user Michael J. McFadden wrote:

You state, "there is reason to believe that there are unknown confounds that may be spuriously suggesting a protective effect of smoking."

Can you expand a bit on what that reason is? I'm guessing you mean there is evidence pointing to such?

Also: I have seen seemingly strong arguments made for Carbon Monoxide blood/cell levels as forming the base of this resistance. Do you have any thoughts on that?

:?<br /> Michael McFadden

On 2021-01-27 16:50:34, user Eric O'Sogood wrote:

A couple things I noticed. Studies that have been peer reviewed and published with large statistically significant effect sizes are reported here as "no data" or, selectively negative individual outcomes from trials which did have positive effect sizes were chosen. I would be interested more in the source of these authors' methodologies. Standardized, widely validated methods were not used here. Considering Kory, Marik et al's meta-analysis has passed peer review and is accepted for publication, and Dr. Hill and Dr. Lawrie, both experienced systematic reviewers for WHO and Cochrane, came to opposite conclusions to these authors, I would say there is an extremely low likelihood this meta-analysis will pass peer review.

On 2021-02-04 13:18:44, user Daniel Hervas Masip, MD, pHD wrote:

It is shocking to observe such a big difference between this meta-analysis and others, For example A. Hills group (https://www.researchgate.ne... "https://www.researchgate.net/publication/348610643_Meta-analysis_of_randomized_trials_of_ivermectin_to_treat_SARS-CoV-2_infection/link/6007a57ea6fdccdcb868a4b3/download)"). It also goes against Tess Lawrie meta-analysis preliminary data. The FLCCC members are not exactly a gang of gangsters; they are serious colleagues. It is starting to be very confusing.

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 2021-01-30 15:14:58, user Joe Psotka wrote:

Autoimmune diseases to increase with covid-19?

On 2021-02-02 03:28:10, user Kenneth Sanders wrote:

Given the prevalence of individuals with previous asymptomatic infection due to SARS-<br /> CoV-2, is there an implication that all individuals (not already confirmed to have had the disease) should be tested for existent antibodies to SARS-CoV-2 prior to first dose of vaccine? Subsequently, only those naive to SARS-CoV-2 before vaccination would receive two doses.

On 2021-02-08 15:48:09, user Thomas McDade wrote:

We just posted this preprint (https://www.medrxiv.org/con... "https://www.medrxiv.org/content/10.1101/2021.02.04.21251170v1)") which documents relatively weak antibody responses in asymptomatic/mild infections in the community--suggests caution in assuming there will be robust immune responses to the first vaccine dose in all seropositive individuals.

On 2021-02-02 22:30:54, user Philippe Marchal wrote:

The authors write "We believe that the large excess mortality seen around the world during the COVID-19 pandemic is robust to the exact model specification". This is clearly false.

Consider for instance controlling for age structure. See

https://www.math.univ-paris...

which is taken from

https://www.ons.gov.uk/peop...

At the end of week 24, there is no excess mortality in France, while the graph on p.5 shows a substantial excess mortality. See also Bulgaria and Czechia, which have a substantial *negative* excess mortality at that time. This does not appear in the graphs.

It should be clear that given the age structure of countries where a baby boom occured after WW2 and given the fact that the mortality rate grows superlinearly as a function of the age, the number of deaths will grow superlinearly in time. A linear regression as used by the authors will not<br /> capture this phenomenon. Thus the related baseline will be lower than the baseline computed by controlling for age structure.

Another major concern is the way the authors modelize the noise $epsilon$ on p.3. I suppose $epsilon$ should be $epsilon_t$, i.e. the noise depends on time, otherwise this makes no sense. But the model seems to assume that the random variables $(epsilon_t)$ are independent, which is obviously not the case: otherwise, there would be no epidemics lasting more than a week! It is a bit ironic that, in a paper studying a pandemic, the authors use a model that cannot describe the annual flu epidemics.

On 2021-02-22 12:53:03, user joe gill wrote:

Thank you for this important report - is there a high resolution version of Table 1 - the linked version does not zoom in clearly to see the country figures. Like a PDF?

On 2021-03-14 19:07:28, user Vikingman wrote:

Hi<br /> Not sure if this will be of any interest but Nicaragua People's Observatory reporting 3002 deaths to Covid related illness rather than the 175 'official' death toll.

https://www.facebook.com/pg...

On 2021-02-04 15:57:58, user JP Monet wrote:

I know that this is in pre-print, but did someone mention that the description of your Group 1, 2 and 3 are inconsistent in your "Methods" section with the description in the Results/Table? This needs to be clarified or it invalidates the conclusions. " Group 1= SARS-CoV-2 IgG negative healthcare worker (HCW). Group 2= asymptomatic SARS-CoV-2 IgG positive HCW. Group 3= symptomatic SARS-CoV-2 IgG positive HCW. Box plots represent 25% to 75% percentile, with individual dots representing outliers using Tukey’s method (1.5 x IQR)." But in Methods, "Group 1: IgG positive with history of symptomatic COVID-19; Group 2: IgG positive and with asymptomatic COVID-19; and Group 3: IgG antibody negative." In this day in age of misinformation, I would want to see your validated raw data to confirm you conclusions.

On 2021-02-06 23:13:06, user sfffff wrote:

Another limitation is the impact of comparing a cohort of non-vaccinated COVID-19 cases to a cohort of influenza cases where approx. 40% of those patients with a higher risk were vaccinated in Switzerland (BAG, saisonbericht-grippe-2019-20.pdf and saisonbericht-grippe-2018-19.pdf). I doubt that this can be included in any sensible manner into the calculations - but it induces a bias, as some of the potentially most critical cases are filtered out (or alleviated). It would be very interesting if you could repeat the study next year, also taking the COVID-19 / influenza vaccinations of your patients into account.

On 2021-02-08 15:46:37, user Werner Bhend wrote:

This study is helpful. But what is unfortunately missing is a detailed age analysis of the hospitalized patients and especially of the intensive care patients. This would allow conclusions to be drawn as to whether herd immunity is really needed or not. Covid 19 is clearly less dangerous for non-risk patients and I would have liked to see a comparison with influenza in the healthy age group 0-65.

On 2021-02-08 16:35:58, user Cristina Aosan wrote:

Thank you so much and congratulations for this great clinical study ! As physician and practitioner of natural therapy I use propolis from 28 years and was sure it acts in Covid 19 infection. The people to whom we recommended after pandemic started, had very good results, both as prevention and as treatment for those already with symptoms, even for those in a severe condition. My surprise was how fast propolis acts. It's excellent to have such scientific confirmation. Thank you again, good luck further on, and waiting other interesting and useful news like this study.

On 2021-02-10 20:18:58, user Isaac See wrote:

The article has been accepted by Clinical Infectious Diseases and can be found (ahead of publication in print) at https://academic.oup.com/ci...

On 2021-02-13 01:49:20, user Julia Alice wrote:

Now published in Analytical Methods!! <br /> https://pubs.rsc.org/en/con...

On 2021-02-13 21:41:54, user Lars Kåre Kleppe wrote:

Very strange description and conclusions<br /> The study must be heavily underpowered. How can a observational period of two weeks in the training arm, where 1/3 either did not attend the centre or maximum two times give meaningful information about exposure in an area where 0,015% of the population tested positive in the actual period. <br /> How can a RT-PCR-testing performed in asymptomatic persons be used to give information about current infection, when PCR can be false positive due to infections that can have occurred sereval weeks in advance AND false negative as they are performed two weeks after they started the training for a disease with an incubation period of up to 10(-14) days. <br /> The study would not be able to properly detect transmission of infection in the second week of the intervention period. <br /> How can a antibody test performed several weeks after the intervention be interpreted for the intervention period alone. After the first wave in Oslo, march-april 2020 the seroprevalence-studies performed varied between 1 and 2 percent and the findings in the study cohort are the same, and can in no way be interpreted as they are.

On 2021-02-16 15:48:05, user Malik Sallam wrote:

This preprint is now published in vaccines journal:<br /> https://www.mdpi.com/2076-3...

On 2021-02-16 19:11:19, user Tim Pollington wrote:

A really relevant study and definitely agree that future modelling should include HIV-VL; in fact reading your other paper I think M/F would be worth including in a mechanistic model too.

Sorry if I misread your paper, however I thought that the main result may not necessarily be that surprising, given that one would expect to find a higher probability of observing presence of PKDL (say) cases at higher VL incidences.

If there was a way of incorporating in your statistical model the current VL, PKDL and VL-HIV counts (as opposed to 'presence of' binary variables) in predicting future VL then could then get at the relative contribution of these three groups, accounting for the infectious time they are around (before treatment or unfortunately their death for HIV-VL patients). I wonder if VL-HIV may be superspreaders wrt the others (as parasite loads would be higher? and not reduce to zero following drugs) which would strengthen your argument re VL-HIV being a forgotten group in VL control.

Tim Pollington.

On 2021-02-17 13:49:48, user David McAllister wrote:

The latest version of this manuscript has now completed peer review and been accepted for publication by the journal Archives of Disease in Childhood.

On 2021-02-18 23:34:23, user Max van Berchem wrote:

There is a mistake in the discussion part. Moderna is the one that had 30 severe cases, all in placebo and Pfizer 9 severe case in placebo and 1 in vaccine arm.

On 2021-02-20 03:35:46, user kdrl nakle wrote:

Community participants? 1540 volunteers and then only 340 completing your study? That sounds to me like a failure of your survey. It is not randomized to start with since you only have volunteers and then most of them did not come back for follow-up so I would assume those that did would be in good health and willing. I think your study is just not anything worth relying on.

On 2021-02-22 08:12:43, user Leo Delibes wrote:

This article has now been published in The Lancet Public Health: https://doi.org/10.1016/S24...

On 2021-02-24 13:41:54, user Nicolas Gambardella wrote:

Figure 1 shows a clear bi-modal distribution of post-infection both at baseline and after one shot. About a third of the patients do not get immunity. It would be interesting to look at the age distribution and time since infection for the two populations.

On 2021-02-28 04:57:17, user Frank Wolkenberg wrote:

It would be very useful to know the criteria for testing. This is not a randomized study, which makes it difficult to understand whether the number of infected cases in the vaccinated sample is equivalent to the number of infected cases in the unvaccinated sample, or whether those individuals represent an anomaly. If this were done, it would help answer the question of to what extent the vaccines protect against infection.

On 2021-03-03 08:20:31, user Martijn Hoogeveen wrote:

Our followup in MedrXiv shows that "Seasonal patterns COVID-19 and Flu-Like Illnesses comparable" https://www.medrxiv.org/content/10.1101/2021.02.28.21252625v1#disqus_thread

On 2021-03-03 16:01:49, user Rafael Onofre wrote:

The article has been published and cab accessed in this link: https://www.sciencedirect.c...

On 2021-03-08 14:55:45, user NickArrizza wrote:

The BIRD meta-anaylsis was an independent review with no conflicts of interest, unlike this one. So is some discernment is required?

On 2021-03-08 16:10:37, user Alberto wrote:

5 out of 25 (20%) healthcare workers in the control group developed COVID-19 vs 0 out of 25 (0%) in the intervention arm.

The number of participants is small, yes. But it sounds like the researchers expected better results! Not very enthusiastic about this small feat.

On 2021-10-19 12:46:17, user student N wrote:

Line 178 & 181, "Supplementary Table 4". You meaned Supplementary Table 5?

On 2021-03-11 20:59:15, user External Reviewer wrote:

The authors recruited their survey participants through a digital campaign on social media. However, the paper does not mention the specific social media platforms or accounts which were used to promote the survey and the potential impact of these accounts on the validity of the study.

The survey was mainly advertised on the first author’s Facebook page which has close to 2.8 million followers and on his YouTube channel which has close to 875,000 subscribers. This explains the massive sample size and the clear bias in the results. In particular, the first author is a well-known science-denier who has been pushing a pseudoscientific creationist perspective onto his social media followers for years.

There is absolutely no surprise that the results came out the way they did. The majority of the survey participants were skeptic of the vaccines and had a general distrusting view of pharmaceutical companies. These people have been conditioned (socially influenced) for years by the first author to believe that the “atheist west” was conspiring against them. At this point, it is almost impossible to convince them otherwise.

Finally, a message to the reviewers, unless the authors provide an evidence to debunk my claims, please do the right thing. This paper should be rejected.

On 2021-03-13 17:03:28, user peterjohn936 wrote:

Improving the Immunity System should have been a standard of care response to COVID especially treatments that are already standard. I am type 2 diabetic, my doctor prescribed vitamin D for me years ago. GPs should be asking their diabetic patients to take extra care to watch their blood sugar levels, to exercise more, and to try to lose weight.

On 2021-03-15 07:47:08, user Pete Austin wrote:

These seem the kind of symptoms that would be caused by living indoors too much, due to inadequate exercise/ventilation and perhaps putting on weight. Publicity for the pandemic will cause people to be more alert to health symptoms, especially following a positive PCR test. I can't find any mention that the same analysis was done for any control group (I searched for "control" and "group"). I wonder if this is a pattern that you would find with other sub-groups of worried well.

On 2021-03-15 13:11:58, user CHPPM TISS wrote:

A cost effectiveness study on an alternative test (FELUDA) for SARS-CoV-2 in India is published and is available at: https://doi.org/10.1016/j.h.... <br /> The FELUDA test has been approved for use in India by its medical device regulatory body and the private healthcare sector in India has started procuring it from the manufacturers.