COVID-19: a visualization of data

From a few reported cases of acute respiratory syndrom in Wuhan (China) in November (?) 2019, the outbreak has rapidly spread to more than 60 countries and/or territories in the world. More than 90,000 cases have been reported in Asia, Australia, Europe, Middle East, Africa, and North America. More importantly, more than 2000 deaths have been recorded worldwide.

There have been numerous studies on epidemiologic characteristics of this epidemic (now known as "COVID-19"). In this note, I try to summarize the data in graphical format so that we can easily understand the magnitude and significance of the epidemic. I will focus on the incidence of new cases, risk of mortality, the number of individuals tested and rate of positives for SARS-Cov-2, coefficient of transmission (R0), incubation period, symptoms and coinfection. I hope that you will find useful information from the figures.

Figure 1: Number of daily (confirmed) cases from January 12 to March 6, 2020. Note that the sudden jump in the number of cases on 25/2 was due to a change in the definition of 'case' by the Wuhan health authority. Unfortunately, they subsequently reversed the definition, and the number of cases was delined afterwards. In recent weeks, the number of cases has increased, due largely to the rise in the outbreak in South Korea, Italy and Iran. Source of data: https://www.worldometers.info/coronavirus.

Figure 2: Number of daily deaths from Jan 12 to Mar 6, 2020. After a gradual decline, the number of deaths appears to have increased in recent weeks due largely to the surge in the mortality in Iran. Source of data: https://www.worldometers.info/coronavirus.

Figure 3: Number of cases (in log scale) stratified by country. Until now (March 6), 96217 individuals from 89 countries/territories have been infected with SARS-Cov-2 virus. Almost 84% of the infected cases were from China. Moreover, South Korea, Italy and Iran have also reported a substantial number of cases. Source of data: https://www.worldometers.info/coronavirus

Figure 4: Relationship between temperatue and the number of infected cases. Countries with moderate or high temperature (eg >20oC) tended to record a lower number of infected cases than countries with mild or cold temperatue. However, this is an 'ecologic correlation', and this does not in any way imply a causal relationship. Source of data: https://www.worldometers.info/coronavirus and my own data collection.

Figure 5: Number of individual tested (upper panel) and the number of positives for SARS-Cov-2 (lower panel) in the UK, Austria, South Korea, USA, France, and Italy. The overall rate of positives was 2.7%. However, this rate varied widely between countries, with the highest detection rate being observed in Italy (5%).

Source of data: https://www.worldometers.info/coronavirus.

Figure 6: The Chinese CDC has published the largest epidemiologic study, in which they reported major characteristics of COVID-19 for 72,314 patients in China. I have used the data to construct this figure. The upper panel shows that a majority (70%) of confirmed cases aged 60 years or younger, while the upper panel shows that most of deaths (81%) [associated with SARS-Cov-2] aged 70 years and older.

Source of data: https://github.com/cmrivers/ncov/blob/master/COVID-19.pdf

Figure 7: I consider that while COVID-19 is still going on, a better way to evaluate the risk of mortality is to estimate the rate of death per 100 person-months. Using the data from the Chinese CDC report, I have constructed the two figures. The figure on the left shows the rate of mortality per 100 person-months by age group, while the one on the right shows the simple case fatality rate (CFR) also stratified by age group. It is not surprising that the risk of mortality increased with advancing age, but the two figures tell different stories. Based on the CFR the overall mortality rate was 2.3% (95% confidence interval: 2.1 to 2.4%). However, the rate mortality per 100 patient-months was 4.64 (95% CI: 0.4 to 8.8).

Source of data: https://github.com/cmrivers/ncov/blob/master/COVID-19.pdf

Figure 8: SARS-Cov-2 is NOT the sole cause of death among infected people. They died because of comorbidities and the infection. This figure shows the risk of mortality per 100 person-months stratified by comorbidity. As can be seen from the figure, patients with existing conditions such as cardiovascular disease, chronic respiratory problem, cancer, type II diabetes have a greater risk of death compared to those without any comorvidity whose risk of 30-day mortality was 1.64. Source of data: "The Epidemiological Characteristics of an Outbreak of 2019 Novel Coronavirus Diseases (COVID-19) — China, 2020", CCDC Weekly / Vol. 2 / No. x.

Figure 9: I guess the rate of mortality associated with SARS-Cov-2 is dependent on the healthcare system of a country. This figure shows the simple/naive case fatality rate by 'major' country. It is astonishing to see the high rate of mortality in the US, but this is likely due to the small number of reported cases. China, Italy and Iran also recorded a higher-than-average rate of mortality. Japan and South Korea appears to have done well!

Source of data: https://www.worldometers.info/coronavirus

Figure 10: This figure shows the incubation period of SARS-Cov-2 and other virsuses that caused SARS, MERS and seasonal flu. It has been reported that the average incubation period of SARS-Cov-2 was 5.3 days, but it could range between 2 ad 14 days (or even longer, according to some studies).

Figure 11: This figure compares the index of transmission (also referred to as 'R0' coefficient) between COVID-19, SARS, and seasonal flu. The estimated R0 for COVID-19 will likely change as more data are accumulated.

Figure 12: Based on data reported in a JAMA paper, I have constructed this figure which shows the pravelence of symptoms among patients infected with SARS-Cov-2. It is not surprising to see that almost 100% patients reported fever, followed by fatigue, and cough. Source of data: https://jamanetwork.com/journals/jama/fullarticle/2761044

Figure 13: Patients are not infected by just one virus; in reality, some patients are infected my multiple viruses. This figure shows the rate of co-infection among patients who were infected with SARS-Cov-2. Source of data: https://www.medrxiv.org/content/10.1101/2020.02.12.20022327v1.full.pdf

What messages can be taken from these data? My personal assessment (and I stress: personal assessment) is that the new coronavirus is much more infectious, but less deadly, than the one that caused SARS about 8 years ago. Although the COVID-19 virus is less deadly, it is associated with far more deaths than SARS (becaus it has affected more people than SARS). I think in the end, we are -- or our immune system is -- going to live with this new virus, because it is highly likely that it will not disappear from the universe. (Bats have lived with this virus for perhaps more than 100 years?)