Anti-HCQ Paper in The Lancet Uses Fake Data

Good news: Early Outpatient Treatment of Symptomatic, High-Risk Covid-19 Patients that Should be Ramped-Up Immediately as Key to the Pandemic Crisis, Harvey A Risch, May 27, 2020,  American Journal of Epidemiology

Five studies, including two controlled clinical trials, have demonstrated significant major outpatient treatment efficacy. Hydroxychloroquine+azithromycin has been used as standard-of-care in more than 300,000 older adults with multicomorbidities, with estimated proportion diagnosed with cardiac arrhythmias attributable to the medications 47/100,000 users … These medications need to be widely available and promoted immediately for physicians to prescribe.

The rest of the post discusses a fraudulent paper (Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis, Mehra et al., May 22, 2020, The Lancet), based on Surgisphere’s fake data, RETRACTED on June 4, 2020.

Surgisphere Corporation is fraud. Some of its services are being used for COVID-19 diagnostics and/or clinical decisions: QuartzClinical, COVID-19 Decision Support Tools (Rapid Triage, Severity Scoring, Diagnosis, Mortality Risk), Surgical Outcomes Collaborative analytics platform.

This anti-HCQ Lancet paper claims: “We included all patients hospitalised between Dec 20, 2019, and April 14, 2020, at hospitals participating in the registry” with COVID-19 and records of death or discharge. 671 hospitals from six continents allegedly provided their data for this database. The US hospitals were allegedly “selected to match the epidemiological characteristics of the US population”. In this great database, the authors have found 96,032 hospitalized patients with COVID-19, including 14,888 treated with HCQ or CQ with either Azithromycin (AZM) or Clarithromycin. 

At this point, one would ask: Who has a database with detailed information on 96,000 patients? What about privacy laws? What about network security and data compatibility? Is Surgisphere a three-letter agency with such unrestricted global access? No, this is a one-man fraud. Paper’s second author is the man. There is no patients’ database and any raw data was fabricated or did not exist at all.

Some of the many mysteries in the alleged data:

  • The population data is very homogeneous, almost the same across the six continents – this is highly unlikely. For example, the patients with qSOFA < 1 are 82.6% in Europe and North America (what a coincidence!), and 82.7% in South America (Table S3).
  • The study data shows only 42% of patients (6,221 out of 14,888) in the treatment groups received HCQ+”macrolide”. However, it is widely known that almost all people treated with HCQ or CQ received HCQ+AZM. 
  • In the study data for North America, 35% of patients (3,415) received CQ and 65% of patients (6,462) received HCQ. However, it is widely known that, outside of China, CQ was used significantly less than HCQ. The actual ratio is closer to 1:99 than to 35:65.
  • All patients treated with CQ/HCQ received antibiotics, almost always AZM. In the study, 32% of the North American patients received no “microlide” (Appendix, Table S3). Do peer reviewers really believe that almost a third of patients hospitalized with respiratory infection received immunosuppressives without antibiotics?  

Tweets from an Australian MD & PhD helped explain some of the mystery. Not only is the data fake, but the database itself, Surgical Outcomes Collaborative, is fake as well. Here are some of the many red flags:

… the study says that they received data from 600 hospitals up to mid-April, and it was published end-May. This is impossible. If you have ever collected clinical research data you will know how impossible it is.

… you need ethical approval, suitable web servers set up, access set up for each hospital (and ideally each worker). This is what the authors say happened – they say they questioned existing repositories

I’m sorry but these don’t exist. Repositories do exist that maintain some of this data but it is mostly unlinked, relies on it being filled in correctly and exists on different systems that don’t talk to each other. …

… for each hospital system providing access you have to go through a complex ethics approval system (otherwise your hospital has just given your personal data to some strangers). …

An ethics approval of this magnitude would take months but it would take even longer to get the system up and running to allow secure access to hospital data systems externally. IT JUST DOES NOT HAPPEN.

The thread continues. The doctor takes note that the patients’ data matches, almost perfectly, on 23 factors.

This is pretty much statistically impossible, and they are claiming that they got this matching in 7000 patients out of a pool of 96000 patients for which they received high quality information from 671 hospitals. Nope. Sorry, the data is too “clean”.


So, this group are telling us that, not only have they got access to 671 hospital clinical record systems but (and this is a problem) … They have access to all the pathology providers linked to those 671 hospitals and THEY CAN DATA LINK THE PATIENT RECORDS. Not a chance. 

This didn’t happen, and if it did it would be a massive privacy scandal. There would be serious financial and potentially criminal repercussions

So, either this data is completely fabricated  -or- The authors have been able to overcome impossible hurdles of data sharing and probably broken a number of laws in the process.

Surgisphere has only a few employees, hired since February 2020. They did not overcome any hurdles. 

This Lancet paper also heavily promotes  Surgical Outcomes Collaborative, the fake database from which the data has been obtained. The “Collaborative” is owned by the Surgisphere Corporation. Sapan S. Desai, one of the paper’s authors, is a Surgisphere founder (disclosed) and CEO (not disclosed). The “Collaborative”, which was unknown before May 2020, is mentioned by name seven times in the paper. Two paragraphs (~500 words) are used as Collaborative advertisement:  “a cloud-based health-care data analytics platform that includes specific modules for data acquisition, data warehousing, data analytics … A manual data entry process is used for quality assurance and validation to ensure that key missing values are kept to a minimum … ensures compliance with the US Food and Drug Administration (FDA) guidance on real-world evidence … data are collected through automated data transfers that capture 100% of the data from each health-care entity at regular, predetermined intervals … the standard operating procedures in place for each of the four ISO 9001:2015 and ISO 27001:2013 certified features of the registry … Collection of a 100% sample from each health-care entity is validated against financial records … capturing all-comer data and consecutive patient enrolment by capturing 100% of the data within electronic systems … compliant with the US Agency for Healthcare Research and Quality guidelines for registries …”.

None of this is true.

Surgisphere Corporation

  • Surgisphere’s website does not display its directors or executives, other than its founder-CEO Sapan S Desai (SSD) (About)
  • The website is excluded from – a sure sign that the company does not want people to know what it looked like in the past.
  • SSD worked in Surgisphere only part-time until February 2020, according to his profile in LinkedIn. His last full-time position was “Full-time clinical vascular surgeon, Medical Director of Surgical Quality, and Director of CME at Northwest Community Hospital” in Arlington Heights, IL
  • LinkedIn also shows profiles of four other employees. The three employees visible to me were hired in February 2020 or later. All of them live in different areas of the USA.
  • The address of Surgisphere is a private house in Palatine, IL, probably SSD’s home. (Buzzfile)
  • Surgisphere is not listed as a tenant at the address listed on its websites (875 N Michigan Ave, 31st Floor, Chicago, IL 60611)

In other words, Surgisphere Corporation existed mostly on paper, until February 2020, when its founder decided to cash out on the COVID-19 epidemic. Its software and services are vaporware.

 The Surgical Outcomes Collaborative (database)

The Surgical Outcomes Collaborative has its own website Surgisphere markets it as an analytics platform, not a database. Nowhere does it  claim possession of or access to patients’ information from multiple hospitals. Further, its website is a re-branding of another company’s website Another Surgisphere’s website is


This Lancet paper does not break down results between AZM and Clarithromycin, nor do they explain why they selected Clarithromycin, so I will. Clarithromycin has harmful interactions with CQ and HCQ (unlike AZM), and Clarithromycin is a strong QT prolonger, (unlike AZM, which is a mild one). This is why physicians did not use Clarithromycin with HCQ as a COVID-19 treatment. I could only find a couple of single patient reports of combining Clarithromycin with CQ. The authors included Clarithromycin in the paper to smear Azythromycin “by association.”


This paper was published on Friday, May 22, on the eve of a long weekend. The paper immediately received wall to wall coverage in the media. On the same day The Lancet published a fawning comment Chloroquine or hydroxychloroquine for COVID-19: why might they be hazardous? These factors suggest careful coordination between the authors, the editors, and news media to catch the public and media community off-guard,causing maximum shock to reverberate over Memorial Dayweekend. 

Surgisphere’s Rapid Triage, Severity Scoring, Diagnosis, Mortality Risk, advertised as artificial intelligence based tools for clinical decision support are nothing more than Javascript code, as simple as an online calculator. Additionally, this calculator is broken. Based on the company’s press releases, these tools have been covered by a range of media outlets from Yahoo! News to Breitbart. They were also covered in professional websites. Wow to those who relied on them. Unfortunately, they are easily accessible on the Surgisphere’s website.

Another Conflict of Interest?

The first author Mandeep R Mehra works in Brigham and Women’s Hospital, which conducts Gilead-initiated clinical trials of Gilead’s Remdesivir for COVID-19.

Published on May 23. Updated on May 24-27, June 4.

Fake Surgispehere’s COVID-19 Clinical Software

Other Bad anti-HCQ Papers

11 thoughts on “Anti-HCQ Paper in The Lancet Uses Fake Data

  1. From the publication it is clear that it is a wrong study. In my years of research and reading I never saw a study done in such a short time. even if it is retrospective. You can’t believe it from there. they say data in late May. when we are still in the month. And to top it off, the WHO believes them assuming that HCQ and AZT do not work. Sure there are other interests. Only time will tell. thanks for the comments.

  2. bravo quelle perspicacité !

    bioformaticien (3 logiciels de gestion d’hopitaux), je confirme toute cela, ca se tient. avant j’avais une impression indescriptible de lapin qui sort du chapeau au bon moment.

    autres infos ici:
    Questions raised over hydroxychloroquine study which caused WHO to halt trials for Covid-19

    en d’autre termes, ils ont construit une jeu de donnée agrégé à partir de sources douteuses :

    Les données sur lesquelles s’appuient les chercheurs pour tirer leurs conclusions ne sont pas FACILEMENT DISPONIBLES dans les bases de données cliniques australiennes, ce qui amène beaucoup à se DEMANDER d’où elles viennent :
    – Le département de Victoria a confirmé que les résultats concernant les données australiennes ne CONCORDAIT PAS avec les données sur les coronavirus de l’État.
    – Le ministère de la Santé a également confirmé qu’il n’AVAIT PAS FOURNI(!!) aux chercheurs les données de ses bases de données.
    – Nous avons examiné notre base de données Surgisphere et découvert qu’un NOUVEL hôpital a rejoint le registre le 1er avril et auto-désigné comme appartenant à la désignation continentale de l’Australasie”,
    -Il a dit qu’il n’avait jamais entendu parler de Surgisphere, et personne de son hôpital, The Alfred, n’avait JAMAIS FOURNI à Surgisphere des données..
    – Habituellement, avec des études qui rapportent les résultats de milliers de patients, vous voyez une grande liste d’auteurs sur le papier”, a-t-il déclaré.
    “PLUSIEURS SOURCES sont nécessaires pour collecter 90000 données et analyser les données pour les grandes études et vous voyez généralement cela dans la LISTE DES AUTEURS.
    -Il a déclaré que l’ensemble de données devrait être RENDU PUBLIC, ou du moins ouvert à un REVISEUR statistique INDEPENDANT
    -Cheng a déclaré que ce serait une erreur d’arrêter des essais cliniques solides et bien conçus examinant le médicament en raison de DONNEES DOUTEUSES.

    en d’autre termes, il n’a que eux qui ont les données, ils ne les fournissent pas en open data. et on doit les croire sur parole !!

    -L’étude ne compte que 4 auteurs, ce qui est étrange pour une étude mondiale sur 96 000 patients (et aucun remerciement à la fin de l’article). Des études comme celle-ci en médecine ont généralement de 50 à 100 auteurs
    – Nous leur avons envoyé un e-mail pour demander les données, en particulier pour examiner l’effet de la dose qui est essentiel pour comprendre les résultats. Ils nous ont dit:
    Merci pour votre demande par e-mail. Nos accords de partage de données avec les différents gouvernements, pays et hôpitaux ne nous permettent malheureusement PAS de PARTAGER des données.
    Données donc indisponibles d’origines inconnues !!!!!
    – Un autre aspect assez remarquable est la BEAUTE uniforme des données agrégées sur tous les continents:
    L’utilisation d’antiviraux est de 40,5, 40,4, 40,7, 40,2, 40,8, 38,4%. Remarquable! Je ne savais pas que le traitement était si bien coordonné à travers le monde.

    “L’étude indique qu’ils ont reçu des données de 600 hôpitaux jusqu’à la mi-avril et qu’elles ont été publiées fin mai. C’est impossible. Si vous avez déjà collecté des données de recherche clinique, vous saurez à quel point cela est impossible”

    je confirme j’ai déjà fait une extraction de donnees, pluiers jours pour récuperer les extraction et les balancer dans une base de donnée. fiches patients tres grosses et tres compleex (sous-entités etc…)

  3. I think 200 mg of HCQ acts as an immunomodulator, but the higher dose ? It is a hazardous one at higher dose .

  4. Another fact that harms the study conclusion ia that its basis includes only patients in advanced stage of SARS COV2 and it ia widelly known that HCQ and CQ are further harmless an efficient If prescribed in the initial stage of the disease.

  5. Mr Goldstein, Your post is interesting. I have no idea if the data (in the anti-HCQ paper) is genuine. But there are some points in your post which I cannot accept immediately.
    I refer to the following comments:
    1. “Who has a database with detailed information on 96,000 patients?” the authors say they took the information from a registry. Registries do exist. These registries anonymize the data, and are accessible to researchers. Data is computerized, and can be quickly analyzed.
    2. “For example, the patients with qSOFA < 1 are 82.6% in Europe and North America (what a coincidence!), and 82.7% in South America." This table has about 30 rows, 6 columns, expressing percentages up to one decimal places. IF the numbers were random, the probability of two numbers in the same row being identical would be 1/1000 x 1/1000, = 0.000001. But the numbers in the same row cannot be random. I mean, the qSOFA for every hospital will be <1 in the vast majority of cases. So the probability becomes closer to 1/100 x 1×100 = 0.0001 if low SOFA scores occur across a 10% band, and even lower if low SOFA scores occur across a narrower range. With 28 such rows, two numbers across different columns will be identical in one or more rows. I'm not a mathematician, but I'm sure we shouldn't calculate probabilities presuming the figures will be random. (Just like the pulse will nearly always be between 60 and 90, not between 0 and 1000.)
    3. "The study data shows only 42% of patients (6,221 out of 14,888) in the treatment groups received HCQ+macrolide. However, it is widely known that almost all people treated with HCQ or CQ received HCQ+AZM.”." From my reading of table S3, CQ alone (1091) or HCQ alone (2127) was given in 3218 cases, while a macrolide was added in 6659 cases, which means that 67.5% of patients did, indeed, receive the macrolide, not 42%.
    4. "In the study, 32% of the North American patients received no “microlide” (Appendix, Table S3). Do peer reviewers really believe that almost a third of patients hospitalized with respiratory infection received immunosuppressives without antibiotics?" The table simply says that the patients did not receive macrolide. It does not specifically state that the patients who did not receive macrolide did not, in fact, receive any antibiotic. In the methods, the authors write that the use of second-generation macrolides was recorded: azithro and clarithro. My own interpretation is that the authors did not record the use of other antibiotics, not that the registry data claimed that no antibiotic was given. Is there some place in the paper that indicates that patients who did not get macrolides got no other antibiotics?
    Sincerely, Suneet Sood

    1. 1. In some countries patients’ registries might exist. Certainly not in the US, EU, or most other countries. Even if such registries exist in some hospitals, why would they give Sapan Desai access to them? There are privacy laws.

      2. The numbers are too uniform and nearly the same in all rows. qSOFA is the most ridiculous one. It can be anywhere from 0.0% to 100.0%, but it is 82.6-82.7 across most important regions!

      3. HCQ+AZM is given almost always. In his time period, it was given in at least 90% of cases.

      4. The antibiotic was almost always AZM because of its synergy with HCQ. Other antibiotics (like Doxycycline) were given in rare instances.

  6. I disagree with this article of the Lancet.
    The WHO could never have base on it, to outlaw its usage.

Leave a Reply

Your email address will not be published. Required fields are marked *