Rushed data collection of suspected early Covid-19 cases in Wuhan

1. The Request for Data Collection

a. Summary:

On Wednesday the 19th Feb 2020, on behalf of the CCP’s national Investigation Team, the “Wuhan Novel Coronavirus Pneumonia Epidemic Prevention and Control Headquarters” sent a “Working Letter”, presumably to the Wuhan Authorities, to ask it to organize some Data Collection from some medical institutions (thereafter “the Data Collection”).

  • For what: on (1) fever cases, (2) imaging of Covid19-like cases, (3) death cases from pneumonia of unknown cause
  • For which period: for medical visits/treatment during the period from October 1st to December 10, 2019. [No reason given for not extending before October the 1st]
  • From whom: at specified medical institutions, with variations depending on the form.
  • How: via 3 forms, to be sent back before Friday the 21st February 2020 5pm [thus only giving one day and a half at most to collect this information].

b. The letter:

The Letter was likely addressed by the ‘Wuhan city novel coronavirus pneumonia epidemic prevention and control Headquarters’ to the Wuhan city authorities, who would relay it to the affected medical institutions.

c. The forms:

Form 1: ‘Fever patients of medical institutions in Jianghan district, Wuhan city’

This form is intended to summarize all medical visits for fever cases from 1st Oct 19 to the 10th Dec 19 within Jianghan District. It applies to:

  • All level-2 or higher medical institutions in the Jianghan District with attached fever clinics (as of the date the request is sent), plus any clinics within three kilometers of the South China Seafood Market (but still in the Jianghan District, so effectively simply lowering the level-2 or higher requirement for that part of the Jianghan District).
Jianghan District and 3km perimeter from Wuhan Seafood Market

Form 2: ‘Wuhan City novel coronavirus like pneumonia cases imaging features ’

The search for cases with novel coronavirus-like imaging features applicable to all medical institutions of level 1 and above in the entire Wuhan municipality.

  • For the level 2 and above medical institutions in the Jianghan district, all cases of novel coronavirus-like imaging in each medical institution are required to be reported (as specified in (1) of the Letter).
  • For the level 2 and above medical institutions not in the Jianghan District, the top 10 cases are to be reported.

Form 3: ‘Deaths from pneumonia in Wuhan city’

Reporting of deaths from pneumonia of unknown cause applicable to all medical institutions in Wuhan City to identify cases with clinical features of the novel coronavirus pneumonia and provide information on each medical record

d. Analysis of the Letter

Confusion and lack of time: As mentioned above the Letter is not that clear at all and would require guidance for the City Authorities to interpret it correctly. Once the intention was clarified, the City Authorities would have then contacted the affected medical institutions. The whole process may have taken hours, limiting the actual time for data processing and case sorting by these medical institutions to about just one day.
On top of that the request letter itself shows signs of having been hastily written. The confusing progression between the collection requirements and their field of application reads a bit like a not yet properly ironed out mash-up of requirements. We also notice a very obvious typo on the 4th line (10⽇1⽇for 10⽉10⽇1) which is rather surprising.

e. The Timeline:

  • On the 31st December 2019, the Wuhan Municipal Health Commission (WMHC) reported a total so far of 27 cases [5] of viral pneumonia, based on search and retrospective investigations of unexplained viral pneumonia cases throughout the city, without giving any earliest onset date.
  • On the 5th January 2020, the Wuhan Municipal Health Commission (WMHC) updated that a total so far to 59 cases [6], and added that the earliest onset date was the 12 December 2019.
  • 23rd Jan 20, the PLA National University of Defence Science and Technology (国防科技大学) jointly establishes with two other universities a coronavirus big-data interdisciplinary research platform. ‘War Epidemic Resumption Big Data Platform’ (战疫复工大数据) uses massive multi-source heterogeneous big data, combined with the occurrence and development of the epidemic, to comprehensively model and analyze the population flow and transmission risk, facilitating government decision-making. [22, 24]
  • 24th Jan, in a widely quoted paper published in Lancet [8], Chinese specialists explain that they found 41 confirmed laboratory cases in patients admitted to hospitals between the 16th Dec 2019 and the 2nd Jan 2020 who were reported by the local health authority. They also report that the first onset case amongst these confirmed cases was on the 1st Dec 2019 [10]. To be clear, the study does not include patients whose specimens were not PCR positive, or outpatients, and crucially has a huge methodological bias against cases before the 15th Dec. Indeed the authors rejected 18 potential cases out of 59 suspected patients, after they did not return positive PCR tests — which in retrospect makes little sense as these tests were done on samples taken around the 4th January 20, which would then typically reject any covid case before the 15th Dec since viral loads would become very difficult to detect that much later (see these notes).
    We note that after this reduction of 59 to 41, the 41 are close to the 44 (mostly suspected) cases reported to the WHO by China on the 3rd January 2020, after the request was made on the 1st Jan 20 (WHO members have two days to answer).[26]
    In any case the study found that 14 cases (especially early ones) had no exposure to the seafood market, which further forced a reappraisal of the South China Seafood Market zoonotic hypothesis, and caused some scientists to suspect a much earlier breakout of October or November.
Extent of early confirmed cases disclosed in Lancet — 24th Jan.
  • 25th Jan, Dr. Daniel Lucey, an infectious disease specialist at Georgetown University, Lucey declared in an interview [14] that if the new Lancet data are accurate, the first human infections must have occurred in November 2019 — if not earlier — because there is an incubation time between infection and symptoms surfacing. If so, the virus possibly spread silently between people in Wuhan — and perhaps elsewhere — before the cluster of cases from the city’s now-infamous Huanan Seafood Wholesale Market was discovered in late December. “The virus came into that marketplace before it came out of that marketplace,” Lucey asserts.
  • 25th Jan, Kristian Andersen, an evolutionary biologist at the Scripps Research Institute, posts his analysis of 27 available genomes of 2019-nCoV on 25 January on a virology research website [15]. It suggests they had a “most recent common ancestor” — meaning a common source — as early as 1 October 2019.
  • 26th Jan, Kristian Andersen and Daniel Lucey positions are reproduced in an article in the well-read Science Magazine. [16]
  • 29th Jan, the Hubei Provincial Committee of the China Democratic League submitted a straight-through proposition “to establish a database of confirmed [and suspected] cases of new coronary pneumonia to support precise prevention and control of the epidemic” to Ying Yong, the Hubei Provincial Party Secretary, who approved it. The suggestion came from Professor Feng Dan, dean of the School of Computer Science of Huazhong University of Science and Technology and member of the Hubei Provincial People’s Government Advisory Committee [20].
  • 24th — 30th Jan, Official Spring Festival Holiday
  • Monday 17th Feb: WHO Full International Team arrives in China as part of the ‘WHO-China Joint Mission on COVID-19'. The full Join Mission Team consists of 25 experts from China, Germany, Japan, the Republic of Korea, Nigeria, the Russian Federation, Singapore, the United States of America and the World Health Organization. The team is co-led by Dr. Bruce Aylward (senior adviser to the WHO Director-General) and co-led by Dr. Liang Wannian (梁万年博士), the head of the COVID-19 response expert team of the National Health Commission of China (NHC).[1, 2, 9, 13]
  • Monday 17th Feb: a paper published in China CDC Weekly gives a lot of information about the Dec 19 cases: 104 confirmed cases, 62 clinically diagnosed and suspected cases. It is not clear whether the paper was discussed with the WHO team at the time, but it was included in its recommended readings on the 19th Feb.
104 confirmed cases + suspected & diagnosed — Dec 19
  • 17th — 22nd Feb: WHO Joint-Mission Team stays in Beijing up to the 17th, then split between a Guangdong Team (18Feb-20Feb) and a Sichuan Team (18Feb-20Feb) before reconvening in Guangzhou. On the program are some visit of various local hospitals, wet markets, transport hubs, government bodies and even the Tencent headquarters in Shenzhen.
  • Wednesday 19th Feb: Data Collection request letter sent on behalf of National Investigation Team sent to Wuhan by Chinese government.
  • Thursday 20th or Friday 21st Feb: Chinese government informs WHO investigation team that they will be allowed to visit Wuhan on Saturday the 22nd.
  • Thursday 20th Feb: an interesting Chinese research paper [27] from the BSL-3 Laboratory of the School of Public Health, Southern Medical University in Guangzhou (previously the First Military Medical University, affiliated to the PLA), is published (in Chinese, on a Chinese site). That paper states that (1) the mean emergence date (tMRCA) of SARS-CoV-2 is around 10th Nov 2019 (with a lower bound for a two-sided 95% confidence interval given as 23rd Sep 2019), (2) ‘natural evolution in nature’ [their exact words] of SARS-CoV-2 from RaTG13 seems difficult to achieve, but evolution is possible from BatCoVZC45 and SARS-CoV.
  • Friday 21st Feb, 5pm: data collected under Data Collection request letter of 19th Feb to be sent back to National Investigation Team.
  • Friday 21st Feb, a very unlikely team lead by conservation biologists from Yunnan publishes a remarkable paper [21], on a Chinese preprint site, that (1) systematically rejects the wet-market hypothesis based on a careful analysis of available sequences to that date, (2) tries to unravel the evolutionary links between the novel coronavirus and RaTG13, (3) complains about the sequences for early cases not being published, (4) clearly states that the breakout may have started in November, not with the official December cases.
  • Saturday 22nd Feb: ‘Select’ members of the WHO Team leave Guangzhou to visit 2 hospitals in Wuhan and have some workshops, spending effectively only one day in Wuhan. According to a New York Times investigation, the team was made of three Chinese and three international experts, and was eventually allowed in after team members agreed the mission would not be credible unless they went to Wuhan. [25]
  • 22nd- 24th Feb: The WHO Team goes back to Beijing, finalizes its reports. The 25 members parsed every word of the report for nearly three days. [25]
  • Monday 24th Feb, the WHO Team holds a press conference and presents its report. Final report available here [2].
  • 25th Feb: On the day following the end of the WHO mission, the CDC issued a gag order signed by 4 top directors of the CDC including Gao Fu. That order affects all CDC offices and stipulates that:
    - There should be no focus on writing research papers.
    - All new research projects must first be validated.
    - No data, sample, or any information on the epidemic can be shared without prior approval.
    - Existing papers submitted but not published must be immediately withdrawn and one should make sure that they comply with the regulation before resubmitting them (if then authorised to do so).
  • Thursday 27th Feb: Pr. Yu Chuanhua [17] is mentioned in the mainland press as working with his team on the processing of a database of 47,000 confirmed medical cases — most likely the official national database of novel coronavirus cases [11] which has been in place since the very beginning of February (based on reporting via the CDC information system), possibly supplemented at this stage by the results of the Data Collection. He only mentions a few possible November cases and no October cases at all, but he does mention a Sep 29 case flagged from the CT scanning (may be in Form 2) from a patient who dies (should be in Form 3).
    The interview seems to have been done on the 25th Feb (day of the CDC gag order) or on the 26th Feb. On the 27th Feb Pr. Yu Chuanhua called the journalist back to retract the two November cases he had just mentioned.
  • 3rd Mar: the State Council ‘Coronavirus Epidemic Joint Prevention and Control Mechanism Research Task Force’ issues a confidential gag order with nationwide effects stipulating that:
    - All publication work is to be taken over by the Task Force which will further ‘guide’ and coordinate the publication of scientific research.
    [‘guide’ is the standard euphemism in China for propaganda work, usually worded as ‘guiding public opinion’].
    - Publication work will be coordinated with the propaganda team of the Joint Prevention and Control Mechanism.
    - All universities, research institutions, medical institutions, enterprises and their staff cannot publish information on the epidemic without approval [effectively a full blanket gag order].
    - Approved publications must be coordinated like a ‘game of chess’ at the national level.

f. Definitions:

Fever Clinic: A fever clinic (发热门诊) is a special clinic set up by the outpatient department of a regular hospital in accordance with superior instructions during the prevention and control of acute infectious diseases. There were 61 fever clinics in Wuhan [3] at the peak of the outbreak and they were officially announced on the 20th January 2020. Each fever clinic reports to one of 9 designated medical institutions (not the same 9 as in form 2).

Hospital classification system of China [4]

2. Results: Early Medical Cases

We have only access to a few snapshots and screenshots of the filled-in forms from the available media reports (Epoch Times and NTD News), not the full forms that are in possession of the media outlets. However based on these few available forms, we can tell that:

a. Form 3: ‘Deaths from pneumonia in Wuhan city’:

The Wuhan Sixth Hospital, one of the nine key hospitals of the Chinese Communist Party, reported five eligible death cases , with two early cases, in the Excel version of Form 3:

Form 3 for Wuhan Sixth Hospital
  • Xu Xgan became ill on October 1 and died on November 3, the cause of death was not specified. Before his death in the Sixth Hospital, Xu Xqian had been given anti-infection treatment in Wuhan Central Hospital.
  • Li Qing X became ill on November 6 and died on December 8 due to severe pneumonia;
  • Another patient became ill on the November 18 and dies on December 5 due to severe pneumonia.
  • Li Xzhen became ill on October 8 last year and died 4 days later (12th) due to a pneumonia
  • Zhou Xlan, became ill on October 14 last year, died on October 31, due to severe pneumonia
  • Xu Xhua became ill on November 10 last year and died 11 days later (21st).

b. Form 2: ‘Wuhan City Covid-19 like pneumonia cases imaging features’ (top 10):

A total 40 cases of fever patients who met the imaging characteristics of novel coronavirus pneumonia-like cases were reported across 8 hospitals alone.

  • Xiao Xgui, the earliest case of the 10, who became ill on September 25
  • 7 people, who all became ill either on the 1st or 2n October, curiously enough.
  • One person who became ill in November 21
Form 2 for Wuhan Puren Riverside Hospital
  • Wu Xyan who became ill on October 3
  • Zou Xlan (Oct 10), Li Xguo (Oct 19), Peng Xiu (Oct 24), Shi Xchen (Oct 30)
Form 2 for Wuhan Yaxin General Hospital

c. Form 1: ‘Fever patients of medical institutions in Jianghan district, Wuhan city:

We have no information on cases reported as per Form 1. These were not discussed directly in the few media reports.

3. Limitations

a. Rushed Data Collection:

It is difficult to imagine that the medical institutions had time to through their admission notes or electronic records to properly fill in these forms. They had only two days at most to do so (Thursday 20th, Friday 21st). In particular the answers to Form 1 (Fever cases) would likely require much more time to collect.

b. No attested Covid-19 diagnostic:

None of these forms is meant to indicate a definitive diagnostic of Covid-19.

  • Form 1 could be expected to cover many different fever pathologies, especially during a normal flu season. However a spike in the number of fever cases may be of interest in the first steps of an investigation, so collecting that information makes sense.
  • Form 2 may point much more directly to suspected Covid-19 cases, especially if filled by doctors who have since then handled many confirmed cases. CT imaging has shown its value in helping diagnostics Covid-19 in patient with fevers [12].
  • Form 3 may also point quite precisely to suspected fatal Covid-19 cases, especially if again filled by doctors remembering the most saillant cases they saw earlier based on recent confirmed Covid-19 cases.

4. Discussion

a. Relation to WHO International Team mission:

The WHO team arrived in China on Monday the 17st Feb. At the time, Dr Tedros Adhanom Ghebreyesus, said that an international team of experts [1] on the WHO-led mission had arrived in Beijing and held their first meeting with Chinese public health officials.

b. Non-Disclosure of the collected data:

In their final report, the WHO mission never mentions any early case or any result of the Data Collection initiative, or that initiative itself. The report actually does not show any case with onsets prior to the 8th December 2019.

Wuhan Confirmed/Diagnosed/Suspected/Asymptomatic by date of onset (test for asymptomatic)

c. Relation to the official national database of cases maintained by Pr. Yu Chuanhua:

It was reported [11, 18] on the 27th February 20 that Pr. Yu Chuanhua (宇传华), vice president of Hubei Health Statistics and Information Society and professor of epidemiology and health statistics at Wuhan University, was working with his team on checking the full database of cases of novel coronavirus (Confirmed/Diagnosed/Suspected/Asymptomatic cases).

Source
National confirmed cases (onset and reporting dates) at the bottom — as of 11 Feb 20
Health Times article on the Analysis of the Wuhan novel coronavirus pneumonia patients database
Nov 14 and Nov 21 cases in actual spreadsheet
  • Why is that official database managed by Pr. Yu Chuanhua, and used in many papers such as [19], not showing the earlier cases that Pr. Yu Chuanhua himself mentions in [11] and [18], including a 2nd Dec onset case which still somehow made its way into the data used by the Lancet study [8]?
  • How can it happen that, after the 21st Feb when the results of the Data Collection were received, the official national database managed by Pr. Yu Chuanhua would still never show any of the suspected Wuhan October 19 cases that were reported, especially the 1st and 2nd Oct onset cases from the Form 2 (CT imaging) of Wuhan Puren Riverside Hospital?
  • Did Pr Chanhua quickly retract the two November cases after learning of the CDC gag order just issued on the 25th Feb 2010?

5. Questions

a. On the WHO visit:

  • Why was the WHO visit of Wuhan delayed until after the completion of the Data Collection?
  • Was the WHO actually shown this data?
  • Was the WHO team directed to these hospitals with early cases?
  • Did China actually fail in its reporting and transparency obligations to the WHO investigation team?
  • What was the real purpose of the Data Collection if it was not used to help with the WHO investigation?
  • Did the WHO cover up this information in their report?

b. On the data itself:

  • How reliable is the data in the forms given the very small time given to process it?
  • How are we supposed to interpret the cluster of imaging cases with similarities to Covid-19 pathology at Wuhan Puren Riverside Hospital with admission dates of 1st and 2nd October 2019?

c. On any follow-up to that Data Collection:

  • Given that the very rushed request for medical and admission data returned some candidates for early Covid-19 cases (going back to the very beginning of October or earlier), did China take the time to do a more thorough and coherent investigation? If not, why? If yes, where are the results?
  • Was the investigation eventually extended to prior to the 1st October 2019?
  • Were these early cases followed to refine the diagnostics, especially in the cases of deaths (for instance by testing any sample for antibodies)? If not, why? If yes, where are the results?

d. On the relation to the official national database of novel coronavirus pneumonia cases managed by Pr. Yu Chuanhua and his team

  • Were the results of the Data Collection added to that national database, even if starting first as suspected cases (especially for Form 2 and Form 3 cases) before further checks?
  • Why is Pr. Yu Chuanhua’s official national database of medical cases (containing confirmed, clinical, suspected, etc, cases) not showing up any of the possible October cases reported through the Forms 2 (which show at least 13 pre-November cases), since he does not mention any suspected October cases and only a few November ones in his interviews in the press only a few days after the closure of the Data Collection ?
  • Were the pre-December suspected cases that Pr. Yu Chuanhua mentioned in his dataset confirmed?
  • In particular was the 29th Sep CT-imaging case he mentioned confirmed?
  • What were the overall conclusions from the analysis and verification work of Pr. Yu Chuanhua’s team of Wuhan pre-December cases present in the official national database?
  • Were these conclusions eventually shared with the WHO?

e. On the relation to the ‘War epidemic resumption big data platform’ managed by the National University of Defence Science and Technology

  • Were the ‘‘War Epidemic Resumption Big Data platform” and its corresponding epidemic data shown to the WHO mission?
  • In particular was there a demonstration of how the big data of population movements during the Spring Festival travel season was used to establish a disease transmission model that was then used to model the spread of the epidemic from Wuhan to all regions of the country? [24]
  • Was the work of Pr. Yu Chuanhua eventually absorbed by the “War Epidemic Resumption Big Data platform”? [22]
  • Why was the “War Epidemic Resumption Big Data platform” with full data resolution never made public?
  • Why was the version of the “War Epidemic Resumption Big Data platform” with limited data resolution, available for a while at the web portal of the National University of Defence Science and Technology (https://nudtdata.com.cn), eventually taken away? [23]

6. Post Scriptum:

This story was initially written on the 6th of October. Since then a few elements have been added, which are worth mentioning:

Confirmation of the Data Collection via WHO conference in Feb 21:

The rushed data collection excercise described here was confirmed by Liang Wannian during the WHO conference on the 9th Feb 21, during which he mentioned it as the ‘Oct — 10th Dec data collection’

Numbers - Terms of Reference of the WHO mission:

On the 5th Nov the Terms of Reference (ToRs) of the WHO mission were quietly published on the WHO website. These terms were negotiated with China and finalised on 31st July. No reason was given for the delay in making them public.

  • Why were these 124 confirmed Dec 19 cases never mentioned before?
  • Was the ‘Retrospective review of cases’ the Rushed Data Collection discussed in this article?
  • Why was this crucial information withheld from at least the 31st July (finalisation of the ToRs) to the 5th Nov 2020 (publication)?
  • Why is the scientific community not given access to these cases and to any corresponding SARS-CoV-2 sequences — absolutely basic steps to do research into the virus phylogeny and its likely origins?
  • Given that there are so many confirmed cases with Dec onset, there must clearly be Nov cases too. Why did the ‘retrospective review’ mentioned not return any? Was it artificially limited to December?

Numbers — ‘CDC Weekly’ paper of Feb 20:

As already mentioned above, the CDC Weekly paper by Zhang & al of the 17th Feb 20 gave a precise count of confirmed cases for Dec 19 in table (104) and approximate counts of clinically diagnosed and suspected cases in a graph (~30).

104 confirmed cases for Dec 19

Numbers — SCMP leak:

An oft cited article in the SCMP of the 13th Mar 2020 gave some numbers from ‘government data’. That was just after the Rushed Data Collection and may well indeed be a first review of that data.

Numbers — CNN leak:

Additionally a leak reported by CNN on the 1st Dec confirms the gist of the article:

Market sampling:

Market samplings did not reveal any animal infected. Only environmental samples (surfaces, waste-waters) showed the presence of SARS-CoV-2.

Gag orders:

After these important papers (such as the Lancet paper showing early cases with no leak to the wet market and and the remarkable paper published on the 21st Feb arguing for more disclosure of early cases and that the outbreak likely started in November at least), the Chinese authorities quickly issued some research gag orders and made sure that the message got through:

Chinese strategy:

A Chinese strategy was then set in place and followed consistently :

  • Do not release any new early cases, hence freezing the picture to early 2020 when the first gag order was issued. The result is that one year later after the disclosure of a total of 44 cases on the 3rd Jan 2020 following an official request by the WHO, there are still only 41 official Dec 19 confirmed cases properly realeased by China (174 if you count the Feb 2021 back-door semi-official release via the WHO), and NO official case before. This despite all the potential for retrospective testing and cases review. This is simply unbelievable.
  • Wait for retrospective sample testings to show early cases in foreign countries.
  • Then point to the US, France, Italy, Spain, etc as the likely source(s?) of the virus and paint China as a victim.

Where do the 200+ cases in 2019 point to?

If the 200 pre-Jan20 cases overall pointed to a foreign import then the best thing that China could do would be to disclose these early cases so as to settle the origin issue at its advantage once for all.

Data Collection — Sources:

  • The original Letter and Forms were made available by the Epoch Times (see articles) and can be found here:
  • The accompanying Epoch Times article has snapshots of the filled forms and provides some more details based on about forms from other hospitals (not in the snapshots).

References & Notes:

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Gilles Demaneuf

Gilles Demaneuf

Opinions, analyses and views expressed are purely mine and should not in any way be characterised as representing any institution.