THREAD (1/10) in which I describe results of preliminary analyses of publicly available #COVID19 data from Singapore. Please interpret these cautiously, as the data are limited and there is no adjustment for confounding, etc.
First, an epidemic curve. No peak in cases yet.
First, an epidemic curve. No peak in cases yet.
(2/10) Next, sex distribution of cases. Males seem to be primarily affected, but these are early days, and emergence of cases in a cluster fashion could be skewing this. Male cases are younger than females (mean 40 vs. 45 years; p=0.164), but this isn't statistically significant.
(3/10) The age of cases has an approximately normal distribution. The mean age of cases is 42 years (standard deviation 14 years). Clearly, all age groups are at risk of infection.
(4/10) 89 cases have been reported to date, all of which I believe have been hospitalised (note this does not indicate the severity of infection). For 77 of these, I could determine the date of admission. The median time between onset of symptoms and hospitalisation was 4 days.
(5/10) 49 cases (55%) have been discharged from hospital, of which I could determine the admission date for 45. For these, the median duration of hospitalisation was 12 days (range 3-29 days).
(6/10) Of 49 recovered cases, 27 are male and 22 are female (note that more males were infected). Of 54 male cases in total, 27 have recovered (50%), and of 35 female cases in total 22 have recovered (62%); p=0.234. Recovered cases are younger (mean 40 vs. 45 years; p=0.065).
(7/10) The data were sourced from the Ministry of Health, Singapore, and analysed by me with #Stata. Once again, I caution that these results should be interpreted carefully. It’s early days and thus based on small numbers, and I might possibly have made mistakes in the analysis.
(8/10) Final disclaimer: These analyses are provided "as is", without warranty of any kind, express or limited.
(9/10) In no event shall the author be liable for any claim, damages or other liability, whether in action of contract, tort or otherwise, arising from, out of or in connection with these analyses or the use or other dealings in the analyses.
(10/10) @joel_mossong @florian_krammer @BogochIsaac @DrEricDing @DrDenaGrayson @gracehwang1970 @COVID2019_ @kakape The data in this thread may be of interest to you.
@joel_mossong @florian_krammer @BogochIsaac @DrEricDing @DrDenaGrayson @gracehwang1970 @COVID2019_ @kakape And most important of all, thank you @sporeMOH for providing these data. These analyses may be of interest to you. Please share with your colleagues in Singapore.
@joel_mossong @florian_krammer @BogochIsaac @DrEricDing @DrDenaGrayson @gracehwang1970 @COVID2019_ @kakape @sporeMOH One more important thing: quite a few people had presented to a GP before ultimately being hospitalised, and some were in quarantine already (e.g., close contact with a known case), so the time between symptom onset and hospitalisation graph should be interpreted cautiously.
@SEACoronavirus @2019nCoVwatcher @CoronaVirusHebr @V2019N The analyses in this thread may be of interest to you. I don't think anyone else has done anything similar yet. Please consider sharing.
