PHWR Public Health Weekly Report

According to Article 2, Paragraph 16, of the “Infectious Disease Prevention and Control Act,” infectious disease surveillance refers to “the entire process of systematically and continuously collecting, analyzing, and interpreting data related to the occurrence of infectious diseases, disease pathogens, and vectors, as well as distributing the results to those in need in a timely manner for use in the prevention and control of infectious diseases” [1]. Generally, surveillance systems are classified into two categories: mandatory and sentinel surveillance systems.

The sentinel surveillance system designates sentinel institutions for infectious diseases under Article 2, Paragraph 16, Subparagraph 2 of the “Infectious Disease Prevention and Control Act” and operates by receiving reports only from such institutions. It is an activity system that involves continuous surveillance in designated institutions with relatively low severity and high frequency of patients, making it difficult to conduct a census. Furthermore, there are cases in which infectious diseases that are unspecified as national notifiable infectious diseases are monitored as supplementary surveillance [2].

The Korea Disease Control and Prevention Agency (KDCA) has been operating a respiratory infectious disease surveillance system divided into clinic-level, hospital-level, and general hospital-level, and has established and is operating a national notifiable infectious diseases surveillance system through influenza-like illness (ILI) surveillance at clinic-level medical institutions, which has been implemented since September 2000, and surveillance of inpatients with acute respiratory infection (ARI) at hospital-level medical institutions, which has been implemented since 2011 [3].

Starting with the surveillance of severe acute respiratory infection (SARI) based on the World Health Organization (WHO) in 2015, KDCA have been implementing a sentinel surveillance system in collaboration with the Korean Academy of Tuberculosis and Respiratory Diseases since 2017, with 42 general hospital-level or higher healthcare institutions (including university and tertiary care hospitals) participating until present [4].

Currently, the sentinel surveillance system for respiratory infectious diseases analyzes data collected from healthcare institutions participating in sentinel surveillance in the integrated epidemic prevention information system and publishes the results through the “Weekly Sentinel Surveillance Report” on the KDCA website for the public, including healthcare institutions participating in sentinel surveillance, local governments, and related ministries.

This study aims to introduce the sentinel surveillance system for respiratory infectious diseases currently operated by the KDCA and present the results of the “Evaluation for Sentinel Surveillance System of Respiratory Virus Infection (2023),” which focuses on the sensitivity and representativeness of the sentinel surveillance system.

In 2023, a policy study was conducted to determine the direction of future development through systematic evaluation of the sentinel surveillance system for respiratory infectious diseases.

Based on the Guidelines for Evaluating Surveillance Systems by the Centers for Disease Control and Prevention (CDC) of the United States, the policy study evaluated the current sentinel surveillance system for respiratory infectious diseases operated by the KDCA in three main categories: performance of the surveillance system, quality of data, and usefulness.

Among them, sensitivity refers to the sensitivity that the data reported through the sentinel surveillance system reflect changes in the overall number of respiratory infection cases in the community. For this evaluation, the data on influenza and respiratory syncytial virus (RSV) reported to the sentinel surveillance system for respiratory infectious diseases were compared with health insurance claims data from the Health Insurance Review and Assessment Service (HIRA). Moreover, data from the HIRA Bigdata Open portal from 2017 to 2022 (influenza disease code J10, RSV disease code J12.1) were used as indicators of the infectious disease population. The analysis was conducted using Pearson’s correlation analysis on the six-year sentinel surveillance report data and the HIRA’s health insurance claim data from the first week of 2017 to the 52nd week of 2022. Using the statistical R program (version 4.1.0), sensitivity was considered excellent when the correlation coefficient was ≥0.8 and p<0.05, adequate when the correlation coefficient was ≥0.7 but <0.8 and p<0.05; otherwise, the sensitivity was considered poor.

Representativeness refers to the accuracy of a surveillance system in describing the distribution of disease outbreaks among individuals and places in a population. In this study, geographical representativeness was assessed by Pearson correlation analysis of the number of surveillance institutions per administrative district with the number of surveillance institutions per population and current number of surveillance institutions. Using the statistical R program (version 4.1.0), representativeness was considered excellent when the correlation coefficient was ≥0.8 and p<0.05, adequate when the correlation coefficient was ≥0.6 but <0.8 and p<0.05, poor when the correlation coefficient was <0.6 and p≥0.05.

1. The Sentinel Surveillance System for Respiratory Infectious Diseases

ILI case is defined as an individual with a sudden onset of fever at ≥38°C accompanied with cough or sore throat [5]. Surveillance is classified as clinical (reporting of ILI case to clinic-level providers) (Figure 1) [6] or pathogen surveillance (collecting and diagnosing the causative pathogen). In 2024, to strengthen sentinel surveillance system for respiratory infectious disease, the number of existing primary care-level sentinel institutions was increased from 200 to 300, including the departments of internal medicine, pediatrics, family medicine, and otolaryngology.

Figure 1. Status of ILI surveillance
ILI=influenza-like illness. Reused from Korea Disease Control and Prevention Agency [6].

Sentinel surveillance for in-patients with ARI has been established in 2011, with 87 tertiary care hospitals and hospitals with ≥300 beds participating in the sentinel surveillance, leading to the establishment of current sentinel surveillance system, including 220 tertiary care hospitals and hospitals with ≥200 beds in 2017, following changes to the designation criteria of sentinel surveillance institutions [7].

Surveillance for ARI includes nine notifiable infectious diseases (adenovirus, human bocavirus, parainfluenza virus, RSV, rhinovirus, human metapneumovirus, human coronavirus, mycoplasma pneumoniae, and chlamydia pneumoniae), influenza, and coronavirus disease 2019 (COVID-19). As the classification of COVID-19 was changed from a class 2 to class 4 infectious disease on August 31, 2023, it was included in sentinel surveillance system from January 2024 after a pilot surveillance period. The number of confirmed cases of notifiable infectious diseases among in-patients at hospital-level healthcare institutions participating in sentinel surveillance is reported weekly through the integrated Information System for Infectious Disease Control (Figure 2).

Figure 2. ARI sentinel surveillance system reporting flow chart
KDCA=Korea Disease Control and Prevention Agency; ARI=acute respiratory infection.

According to the current WHO, SARI case definition is that patients with ARI who have history of fever (or measured fever of ≥38℃), cough and onset within the last 10 days (symptoms within 10 days) and require hospitalization [8]. The surveillance has been operated mainly during winter through a pilot project of a hospital-based surveillance system model involving nine healthcare institutions in 2006 and expanded to 16 and 22 healthcare institutions in 2015 and 2016, respectively, to establish an sentinel surveillance system for SARI. Thereafter, to strengthen the sentinel surveillance system for respiratory infection diseases following the COVID-19 pandemic, the number of participating institutions increased in 2020, and surveillance was implemented year-round, leading to the establishment of the current sentinel surveillance system with 42 tertiary care and university hospitals. For the reporting method, patients who have been admitted in tertiary care and university hospitals participating in the sentinel surveillance and met the criteria for SARI are reported weekly through the Intergrated Information System for Infectious Disease Control (Figure 3).

Figure 3. SARI sentinel surveillance system flow chart
KDCA=Korea Disease Control and Prevention Agency; SARI=severe acute respiratory infection.

2. Evaluation of the Sentinel Surveillance System for Respiratory Infectious Diseases

The overall sensitivity of the sentinel surveillance system for respiratory infectious diseases from 2017 to 2022 was deemed excellent (Table 1).

Table 1 Respiratory infectious disease sentinel surveillance systems assessment results

Evalution indicator	Evalution target	Detailed evalution indicator		2017	2018	2019	2020	2021	2022
Quality of data	ILI	Sensitivity		Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
		Representativeness		Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
	ARI	Sensitivity	Flu	Excellent	Excellent	Excellent	Excellent	Insufficient	Excellent
			RSV	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
		Representativeness		Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
	SARI	Sensitivity	Flu	Excellent	Excellent	Excellent	Excellent	Insufficient	Excellent
			RSV	Excellent	Excellent	Excellent	Excellent	Excellent	Excellent
		Representativeness		Excellent	Appropriate	Insufficient	Appropriate	Appropriate	Appropriate

ILI=influenza-like illness; ARI=acute respiratory infection; SARI=severe acute respiratory infection; Flu=influenza; RSV=respiratory syncytial virus..

In the surveillance system for ILI cases, the overall sensitivity was excellent, with correlation coefficient >0.9 and p<0.05 in all years of the evaluation period.

In the sentinel surveillance system for ARI, the overall sensitivity during the COVID-19 pandemic was considered excellent, with correlation coefficient >0.9 and p<0.05 in all years, excluding data for 2021. The correlation coefficient for influenza 2021 was poor at 0.237 with p=0.458, while the correlation coefficient for RSV was excellent at 0.992 with p<0.05.

The analysis of influenza in the early stages of the COVID-19 pandemic showed that there were few reports in the surveillance system due to the tightening of social distancing.

In the sentinel surveillance system for SARI, the correlation coefficients were >0.8 in all years, excluding data for 2021, and all p<0.05, indicating excellent sensitivity. In 2021, the correlation coefficient for influenza was 0.473 with p=0.12, indicating poor sensitivity, whereas the correlation coefficient for RSV was >0.9 in all years with p<0.05, indicating excellent sensitivity. Therefore, the overall sensitivity for the evaluation period was considered excellent.

Geographical representativeness was considered excellent in the sentinel surveillance system for respiratory infectious diseases from 2017 to 2022 (Table 1).

In the surveillance system for ILI cases, it was evaluated based on the geographic location of influenza sentinel institutions and number of influenza cases per administrative district, all correlation coefficients ≥0.9 and p<0.05 in all years of the evaluation period, indicating excellent geographical representativeness.

Considering approximately 200 sentinel institutions, the overall distribution of sentinel institutions during the evaluation period was considered excellent, but if observing each administrative district (city and county), there were disparities in the number of designated sentinel institutions (either lacking or excessive); Therefore, it is recommended that the number of sentinel institutions be adjusted to resolve these imbalances.

In the sentinel surveillance system for ARI, the correlation coefficient was ≥0.9 and p<0.05, showing significant results; therefore, the overall representativeness was considered excellent.

However, the representativeness was relatively lower in the sentinel surveillance system for SARI than in other surveillance systems. Over the evaluation period, the geographical representativeness was excellent in 2017 (correlation coefficient ≥0.8, p<0.05) but was poor in 2019 (correlation coefficient ≥0.4, p<0.05). This may have been due to the insufficient number of sentinel institutions in Gyeonggi-do in 2019. The geographical representativeness was adequate (correlation coefficient ≥0.6, p<0.05 in all years, except 2019).

These findings suggest that the current sentinel surveillance system for respiratory infectious diseases has provided a sensitive and representative view of the incidence and prevalence of respiratory infectious diseases in the community, excluding the COVID-19 pandemic.

The Republic of Korea (ROK) has been operating its sentinel surveillance system for respiratory infectious diseases according to the WHO recommendations, and policy studies have confirmed that the sentinel surveillance system provides a sensitive and representative view of the incidence and prevalence of respiratory infectious diseases in the community. However, this study has limitations: data provided by HIRA was utilized due to data acquisition for sensitivity evaluation, and since the evaluation period was from 2017 to 2022, the survey was conducted through relied on the respondent’s recall, requiring caution in interpreting annual evaluations.

The KDCA will work to further enhance the representativeness of the surveillance system for respiratory infectious disease outbreak patterns in the community and to expand surveillance institutions to produce surveillance results for community epidemic patterns in addition to producing nationwide surveillance results.

Additionally, considering that all previous epidemics of emerging infectious diseases have been respiratory infectious diseases and the shortening of the epidemic cycle, a reliable representative surveillance system for the outbreak of respiratory infectious diseases in ROK is necessary. The KDCA will continue to identify areas that need improvement through regular evaluations of the sentinel surveillance system for respiratory infectious diseases and further develop the surveillance system through communication and cooperation with sentinel institutions.

Ethics Statement: Not applicable.

Funding Source: None.

Acknowledgments: Thank you for Professor Sukhyun Ryu and the research team at Konyang University and the other university, institute researcher who participated in the study.

Conflict of Interest: The authors have no conflicts of interest to declare.

Author Contributions: Conceptualization: HRC, JHH, IHK, JSY. Data curation: HRC, YJS. Investigation: HRC, YJS. Formal analysis: HRC, JOC, YJS, JHH, IHK, JSY. Project administration: HRC, JOC, JHH, IHK, JSY. Resources: HRC, JOC, YJS, JHH, IHK. Supervision: HRC, JHH, IHK, JSY. Visualization: HRC, YJS. Writing – original draft: HRC, JHH, IHK. Writing – review & editing: HRC, JOC, YJS, JHH, IHK, JSY.