COVID-19 in South Korea - and potential implications for us (wherever we are)

(Updated: 2020-10-03 21:34:27 EDT)

So, I have been watching the epidemic in China and Korea since the beginning, but it is still not easy to settle into this reality here in the US.

First of all, everyone, STAY HEALTHY and see what YOU can do!

Second, there are many excellent news articles and information sources, but I’m sharing some more questions (see left panel) and answers for those who are curious about the situation in South Korea, specifically. Please note that the country has had a trace-test-isolate strategy consistently from the outset of this epidemic. If you want to know more about South Korean situation - compared to other countries, see this Washington Post article: How countries around the world have tried to contain the coronavirus

All data are from press release from KCDC - available to the public in both Korean and English. Kudos!


Q 10. From “social distancing” to “distancing in daily life”: keeping an eye on new cases

The first wave in South Korea has been completed unequivocally. On May 6th, the government announced a major pivot in its COVID19 control, to focus on prevention of transmission, including “distancing in daily life”, and transition to new “daily life” (not the typical normal life before the pandemic though). It will be important to keep the level of infections under a threshold to prevent, delay, and control the next wave.

The following figure shows a trend of new cases (gray bars), with a 7-day rolling average (black line). The horizontal dotted line is the number of new cases on February 23rd, when the cumulative incidence exceeded 1 per 100,000 population (i.e., 154 cases, averaged over 7 days) and the new infections increased exponentially.

Nevertheless, a new cluster of transmission was identified over a popular family holiday period. The government is doing extensive tracing and testing, repeating its earlier strategy. We will keep a close eye on the trend of new cases and new tests (see Question 5 below).

Hover over each figure to see values and more options.

(Source: KCDC’s daily press release on 2020-10-03)
Note: Three phases (red/orange/yellow shades) based on the trend of new confirmed cases. See here for more information on that classification.


Q 9. How are different regions affected in the country?

This is one of common questions that I’ve received, as we see huge variation in incidence rates across states and counties in the US.

In South Korea, however, the entire country is under one same strategy for infectious diseases control, even though the country is decentralized with nine provinces and eight cities. So it’s not surprising that there is practically no regional differences in terms of the incidence or case fatality rates, except Daegu city and Gyungbuk province (a province surrounding Daegu), which had a massive outbreak related to a religious sect in late February. Until mid August, still a majority of total confirmed cases (about 60%) have occurred in Daegu and Gyeungbuk province. However, outbreaks in Seoul/Gyungi in mid August may shift the distribution.

  • In Gyungbuk, including Daegu city, 140 people in 100,000 population have been infected.

It is notable that the outbreak in Daegu and Gyungbuk did not spread throughout the country more widely, although the country is relatively small - about the size of Kentucky or Indiana in the US. For further regional data, see here.

Hover over each figure to see values and more options.

(Note: special cities are included in provinces where they are located: ; Daejeon and Sejong in Chungnam province; Seoul and Inchon in Gyeonggi province; Daegu and Ulsan are included in Gyungbuk province; Busan in Gyungnam province; and Gwangju in Jeonnam province. For case fatality rate, Gangwon has a small number of confirmed cases and was further combined with Gyungnam.)


Q 8. Is it spreading faster in some age groups than others?

So, we know about the age-pattern of mortality (i.e., higher in older ages, see Question 4) and relatively young patient population in South Korea (See Question 3). Meanwhile, as I saw overall mortality going up steadily (See Question 7.1), I wanted to know if COVID is spreading more rapidly among those who would have higher mortality risk (e.g., older people). One way to answer that question is to see if infection incidence rate (number of confirmed cases per population) has increased more rapidly in one age group than others.

In all age groups, there was a rapid increase in early March, when the country had a huge and acute outbreak (sorry, it’s not very clear in the figure, because the age-specific trend data starts from March 3rd). Following that, however, the infection rate has increased more rapidly among those who are 80 and older than in any other age groups (red line in the below figure): from 5.8 per 100,000 people on March 3rd when detailed data by age were released first, to 20.5 on March 21st, and 52.9 on 2020-10-04. The increase around mid-late March reflects outbreaks in nursing homes.

But, it is also notable that infection rate among the 20s has increased steadily since late March - which may be due to ‘imported cases’ among Koreans returning from overseas (this can be confirmed if age breakdown among those imported cases is available).

Hover over each figure to see values and more options.

(Source: KCDC’s daily press release, and UN World Population Prospects 2019 Revision)
Note: Age-specific data were first released on March, 3rd, and it has been updated daily since March 11th.


Q 7.1. How has mortality changed over time?

South Korea had the first COVID death on February 19th. As of 2020-10-03, there have been a total of 420, with an overall case fatality rate of 1.7% among all cases.

The below graph shows trend of the overall case fatality rate (starting from February 21st). The case fatality rate (orange bars) initially increased for several days, but, as the number of new cases increased exponentially (see gray bars also Question 2), the case fatality rate declined after the first peak. Then, it has increased gradually to the current level. Since South Korea still has been able to continue the trace-test-isolate strategy, it will be important to monitor the trends, providing important insight about this epidemic.

Hover over each figure to see values and more options.

(Source: KCDC’s daily press release, and UN World Population Prospects 2019 Revision)


Q 7.2. How has mortality changed over time, by age group?

The graph below shows trends of mortality among those with confirmed COVID-19 by age group - starting from March 3rd. Case fatality rate among those 80 and above has increased continuously. As of 2020-10-03, 21.385% of people in their 80s who have confirmed COVID have died, increased from 5.556% on 2020-03-03. This increase is not surprising, considering the typical clinical timeline and prognosis of COVID infection. A more important question is if and when the case fatality rate will level off, as this epidemic progresses and a bigger proportion of population is affected.

Case fatality rate among the 70s has also increased but at a slower margin. Case fatality among those in their 50s and 60s have remained similar over time or increased slowly.

Hover over each figure to see values and more options.

(Source: KCDC’s daily press release, and UN World Population Prospects 2019 Revision)
Note: Age-specific mortality data were first released on March, 3rd, and it has been updated daily since March 11th.


Q 6. How is it affecting children?

Compared to other age groups, incidence rates are much lower among those under 20 (See Question 3). There have been 0 deaths reported in these age groups, and case fatality rates are also the lowest among all age groups (See Question 4).

Still, new detailed data from Korea CDC give some information about cases among children younger than 18 years of age. As of 2020-03-17, incidence rate by detailed age group is shown below. Older children have higher infection rates.

Hover over each figure to see values and more options.

(Source: KCDC’s daily press release on 2020-03-17, and UN World Population Prospects 2019 Revision)
Note: These age groups roughly corresponds with school system age groups in South Korea.

Of course, this is only epidemiologic perspective. Children’s emotional, social, and economic well-being is more complex and difficult to capture yet.


Q 5. How extensive has testing been?

Very aggressive. As of 2020-10-03, a total of 2340 thousand people have been tested. There are 51 million people in the country, and 46 in every 1000 people have been test. This is one of the highest cumulative testing rates in the world. Epidemiological investigation of the case No. 31 (who was confirmed on February 18th) prompted immediate and massive testing in Daegu and nation wide.

In early May, a new cluster of transmission was identified after a period with a relatively small number of cases. The number of new daily tests increased substantially back to the level in late Februray and early March, following extensive tracing related to the outbreak.

Importantly, except in early March, the rate of positive results out of all tests has remained low - i.e., substantially below 10%. The black line below is 7-day rolling average of daily positive rates.

(Source: KCDC’s daily press release. The number of tests is calculated based on the daily increase in the total number of tests reported.)


Q 4. What is the mortality - overall and by age group?

Among those who have COVID, mortality rate among the elderly with COVID is very high: 21.4% for those who are 80 or older, and 7.3% for those who are between 70-79.

Again, this is the case under a very aggressive trace-test-isolate strategy, but with no nation wide lock down. As of 2020-10-03, a total of 420 people with COVID have died, 1.7% among all COVID cases.

There are news articles about extremely over-burdened health systems especially in Daegu. I’ve read a few mortality case on the newspaper while he/she was waiting for hospital admission, and patients from Daegu and a surrounding province being transferred to other less-affected provinces. However, my impression (as a public health trained reader) is that most severe cases do receive treatment at hospitals. I’m looking for more systematic data on health systems’ capacity and response.

(Source: KCDC’s daily press release on 2020-10-03, and UN World Population Prospects 2019 Revision)


Q 3. What age groups are affected?

For this question, let’s look at the total cumulative number of confirmed cases since the outbreak per 100,000 population by age group, not the absolute number. Also, keep in your mind, this is what we see under a very aggressive trace-test-isolate strategy (see this and this). Finally, KCDC releases the number by 10-year age group, and I’m following that categorization.

  • Incidence rate is substantially high in the 20s: 71.2 people have been infected per 100,000 population in their 20s. At least partially, this is reflection of a very concentrated epidemic in Daegu related to a church, and the age pattern among the church attendants. So this age pattern may not be relevant for more generalized pandemic situation.
  • Incidence rate among those under 20 is low. It may indicate either children are less susceptible, and/or - more importantly - they have been protected at home since all Korean schools have been in long winter break (typically over several weeks) or under temporary closure since the end of December, 2019.

(Source: KCDC’s daily press release on 2020-10-03, and UN World Population Prospects 2019 Revision)


Q 2. How fast has it been spreading?

For this question, let’s focus on the number of new cases by day. For about a month since the first case, only 30 cases were confirmed - mostly in/near Seoul. However, starting from the case No. 31 (a super transmitter who was confirmed on February 18), the number of new cases exponentially increased - primarily in Daegu and a region surrounding it but throughout the country. For 10 days, more than 400 new cases were confirmed everyday, with a peak of 909 new cases on February 29. Since the peak, the number of new cases decreased consistently, and the first wave has been completed in South Korea. This question has been merged with the new Question 10.

As of 2020-10-03, a total of 24027 people have been infected. It means 47 people have been infected out of 100,000 population since beginning of the epidemic.


Q 1. When did it start?

On January 20, 2020, there was the first laboratory confirmed COVID case.


More questions and data coming soon, as I continue digging the awesome KCDC website…
See GitHub for data, code, and more information.
For typos, errors, and questions, contact me at www.isquared.global.

Making Data Delicious, One Byte at a Time, in good times and bad times.