Analysis results of 25,000+ lab-confirmed COVID-19 cases in Wuhan till Feb 18: epidemiological characteristics and non-pharmaceutical Intervention effects. The key results are summarized below and hope they are useful for the strategies in US and others medrxiv.org/content/10.110…
1. In the absence of intervention (before 1/23), the effective reproductive number R=3.8>>1, and dropped to R=0.3 after the massive intervention on Feb 1(centralized quarantine and treatment of all confirmed and suspected cases). This intervention worked.
2. Various sensitivity analyses were done and the conclusion holds.
3. Family transmission is common. Quarantine-at-home between 1/23-2/1 helped and did not work effectively (R=1.35>1), as family members and close contacts may be infected and might affect the neighborhood communities, e.g., going shopping. Centralized quarantine is critical.
4. About 60% of infected cases were estimated to be un-ascertained, i.e,. they scattered in the community and could infect others. This means increasing the testing capacity for early diagnosis is critically important.
5. Healthcare workers and elderly people are at a much higher risk of being infected. The risk of severity increases with age, and elderly people are at a much higher risk of becoming severe (RR=3 for 60-79 and RR=5 for 80+).
6. Women have a lower risk of severity compared to men (RR=0.9). Early diagnosis and early treatments are important.
7. Children's infection risk is much lower than adults, but the risk increased with time.
8. Special efforts are needed to protect vulnerable populations (healthcare workers, elderly people and children).
Thanks to the hard work of all my co-authors at Tongji SPH in Wuhan on finishing this paper in a short time window. The gained knowledge benefits US and other countries in our strategy development. It was a great honor to work with them
Thanks particularly go to Chaolong Wang*, Li Liu*, Xingjie Hao*, Huan Guo* (co-first authors) and An Pan, Sheng Wei, Tangchun Wu (co-last authors) at Tongji School of Public Health in Wuhan for their tremendous work and effort.
Huge thanks also go to all the Wuhan residents, the local health professionals, and the 40,000 healthcare professionals cross the country who went to help Wuhan, for their tremendous sacrifice and efforts.
In this local talk, I gave the background, summarized and interpreted the findings. I provided several take home messages and practical recommendations for the current situation in US. hsph.me/covid19. Updated slides: hsph.me/27v #COVIDー19 #coronavirus 
Very sad that BostonGlobe reported a Brigham and Women Hospital doctor was infected yesterday. Very important that healthcare workers are fully protected. See my slides 56 and 38 at hsph.me/27v for recommendations on what needs to be done for their protection
Starting the effective intervention, centralized quarantine of ill patients and exposed subjects ASAP, will significantly reduce the number of new infections, stop the outbreack, save lives, and relieve the enormous pressure on health system, eg a lack of ICU capacity
It will protect our beloved family members from being infected, help patients receive immediate treatments early, reduce community transmission, reduce mortality by preventing progression to severity, which has a much higher risk of death.
So all children can still have parents and grandparents. Need culture adjustment. It is not easy and will cause short term inconvenience for many of us. But a short term suffer will buy a big long term gain for each of us, the community, the world, and our children.
Please help talk with your institution, community, local, state and federal government leaders. We need their help.
Compared with Wuhan which started the intervention when a very large number subjects were ill and exposed, it is much more easier and effective to do this given US and many other countries are still at the early stage and the numbers of ill patients and exposed subjects are small
I gave an updated zoom talk on "Learning from 26,000 COVID-19 cases in Wuhan" at the Broad Institute on Friday, March 20. Here is the recording. It provides several updated take home messages and priorities.
Here are the updated Broad talk slides. docs.google.com/presentation/d…
Several people asked the reason of the outlier on Feb 1 in the daily count histogram. During the lockdown period, there was a serious shortage of beds and HCWs. An overwhelmingly large number of patients could not get medical care.
In response to this, a few field hospitals were launched and planned to be open on Feb 1. Staff tried to get a very large number of people admitted to the field hospitals ASAP on Feb 2, and did not have time to record the exact diagnosis date for each patient.
They simply filled the diagnosis date as Feb 1. The true diagnosis dates for these cases in that outlier should be a little earlier than Feb 1.
