4 DISCUSSION
Since the first case of COVID-19 in the country on January 20, 2020, South Korea performed aggressive screening tests to ensure early detection of individuals who developed the infection and traced, quarantined, and treated patients as necessary, based on which it was considered to have done well in COVID-19 control compared to other countries. Aggressive COVID-19 testing was performed via Real-time RT-PCR, and the Ct value, which is the number of amplification cycles performed to cross a certain threshold, is used to determine positivity. This value can also be understood as viral load.
Recently, studies have expanded beyond the dichotomous use of Ct values for determining COVID-19 positivity or negativity and began to explore Ct values in relation to demographic characteristics, disease severity, and mortality rates. In addition, some studies have investigated the use of Ct values for predicting future epidemic trends and monitoring infectious diseases. Therefore, the present study aimed to examine the time series distribution of actual Ct values using nationally collected data and analyzed its association with the number of new cases and Rt trends to discuss the utilization of laboratory-based COVID-19 test results.
From January 20, 2020 to January 31, 2022, a total of four COVID-19 waves occurred in Korea. Hence, we analyzed the associations among daily Ct values, number of new cases, and Rt for the entire COVID-19 period and each COVID-19 wave. During the entire period, Ct values generally decreased, indicating an increased viral load over time. In relation to the number of new cases, the number of new cases surged after a substantial drop in Ct values. This reflects the fact that the number of new cases increases markedly after viral load is relatively increased. In terms of the different periods for COVID-19 waves, the average daily number of new cases was markedly higher during the second wave (142.8) than first wave (71.5), despite the fact that the cumulative number of cases was similar between the two periods, showing that the virus spreads more in a shorter period of time when Ct values are low.
When compared with Rt, which is used to predict incidence, the time series trends of Rt and Ct values were similar. However, when compared to the actual number of new cases, Ct values predicted new cases earlier than Rt. Moreover, Ct values were more sensitive than Rt in predicting new cases before the later COVID-19 waves when the incidence rate skyrocketed.
We also compared Ct values against the emergence of variant viruses and launching of vaccination, which may contribute to changes in Ct values, but there were no significant changes in Ct values before and after these periods. The Alpha variant, discovered in late December 2020, had been the dominant variant until mid-July 2021, and the Delta variant became the dominant variant from July 2021 to end of December 2021. The Omicron variant was discovered in late November 2021 and became the dominant variant since. Ct values hit bottom in early December, before the emergence of the Alpha variant, and rose drastically until mid-January 2021. Then, Ct values again decreased until early July 2021, when the Delta variant emerged. Since the emergence of the Delta variant, Ct values remained high without marked changes. Although Ct values began to drop after February 26, 2021, when the vaccine rollout was begun, but the data shows that the timing of increase began to decline from before the initiation of vaccine rollout.
Our findings from the comparison of time series trends of Ct values obtained by analyzing nationwide data with trends of new cases show that Ct values can be used to predict future incidence rates and size of epidemic. Moreover, Ct values were found to predict the size of an outbreak more quickly than Rt, the classic predictor of epidemic trends. Furthermore, Ct values fluctuated more than Rt during the early days of the pandemic before the emergence of variants of concern, such as Delta and Omicron variants, highlighting its potential as an indicator sensitive to incidence rates. One limitation of this study is that our study data—although it was nationally collected data—only accounted for 40% of the total data and lacked test results for two out of 17 cities and provinces in Korea. However, there is a low risk of bias, as the data was collected through the same testing and sampling methods in a single diagnostic laboratory.
The findings show that Ct values can be used in the prediction of future outbreaks and determine viral load. Thus, these findings will serve as evidence supporting the use of Ct values to assist in public health decision making to determine the degree and timing of infection control responses as well as resource allocation, beyond simply using them for COVID-19 diagnostic purposes.