Of the 5 types of viral hepatitis identified to date, hepatitis B is associated with the highest global disease burden. The World Health Organization (WHO) estimates that approximately 296 million people, or 3.8% of the world’s population, are living with chronic hepatitis B infection. Additionally, about 1.5 million new infections occur annually [1]. In response to this situation, the WHO has continued its efforts to eliminate hepatitis. In 2016, the World Health Assembly adopted a resolution establishing July 28 as World Hepatitis Day. The goal is to achieve a 90% reduction in new cases of viral hepatitis and a 65% reduction in mortality by 2030, relative to the figures from 2015 [2,3].

In 1982, prior to the introduction of the hepatitis B vaccine, the prevalence of hepatitis B surface antigen (HBsAg) positivity in the Republic of Korea was 8.6%, constituting one of the highest prevalence rates in the world [4]. However, following the introduction of the hepatitis B vaccine in 1983 and the implementation of a national vaccination program in 1995, the prevalence of HBsAg positivity for hepatitis B in this country significantly decreased, dropping to 2.0% by 2022 [5]. Despite this progress, the socioeconomic burden of hepatitis B remains substantial in the Republic of Korea. Mortality statistics indicate that hepatocellular carcinoma (HCC) and liver disease represent the fifth leading cause of death in the country, after cancer, heart disease, pneumonia, and cerebrovascular disease [6]. Furthermore, an analysis of data from the National Health Insurance Service suggests that hepatitis B accounts for approximately 60% to 70% of HCC cases [7].

The Republic of Korea currently employs 2 methods for monitoring hepatitis B outbreaks. The Korea Disease Control and Prevention Agency (KDCA) has designated hepatitis B as a class 3 notifiable infectious disease and maintains a surveillance system for the condition. Additionally, the Korea National Health and Nutritional Examination Survey gathers annual data on the HBsAg positivity rate and publishes information on the prevalence of hepatitis B within the country [5,8]. Due to a mandatory vaccination program for infants and young children, along with the Perinatal Hepatitis B Prevention Program, the incidence of cases via vertical transmission—historically the primary mode of infection—has markedly decreased. However, horizontal transmission of hepatitis B continues to occur [4]. Consequently, surveillance efforts are exclusively focused on acute cases to better understand the patterns of horizontal transmission of the disease.

Acute hepatitis B is defined as an infection that occurs within 6 months of exposure. It is characterized by cold-like symptoms, including nonspecific gastrointestinal issues such as anorexia, nausea, vomiting, and lethargy, as well as jaundice and systemic symptoms. Notably, up to 60% of infections may be asymptomatic [6]. Since its inception in 2000, hepatitis B surveillance in the Republic of Korea has undergone changes in methods and targets; currently, only acute hepatitis B is monitored. It is therefore imperative to assess how these changes have impacted the effectiveness of hepatitis B surveillance and to ascertain whether the system is functioning as intended [9]. In this study, we also evaluated how the present surveillance system for acute hepatitis B contributes to the goal of eliminating viral hepatitis and identified potential improvements aligning with the national strategic plan for this disease.

Data in the form of 5,117 reports from medical institutions and 3,097 case reports from the KDCA’s mandatory surveillance system for acute hepatitis B were analyzed over a 10-year period, from 2013 to 2022. To assess the surveillance system, 3,142 records of patients who were ultimately included in the statistics were matched with their corresponding case reports. To evaluate the positive predictive value, report data from medical institutions that were not included in the statistics—due to not meeting the diagnosis and reporting criteria—were considered. The analysis was conducted using Excel ver. 2016 (Microsoft Corp.).

We utilized the Updated Guidelines for Evaluating Public Health Surveillance Systems, provided by the United States Centers for Disease Control and Prevention (CDC) [10]. Our evaluation focused on 5 key attributes: simplicity, positive predictive value, data quality, timeliness, and usefulness.

Simplicity is a metric that assesses the structure and ease of operation of a surveillance system. The more straightforward the system, the higher its rating. The evaluation of simplicity encompasses the data reporting system and structure, the data collection process, and the case investigation procedure. The specific criteria necessary for reporting diagnoses were considered in our evaluation. Additionally, the potential for improvement in the diagnosis and reporting criteria for hepatitis B within the existing surveillance system was outlined.

Positive predictive value is an indicator of the proportion of reported patients who are confirmed to be actual patients, reflecting the accuracy of a surveillance system in detecting true cases of a disease. However, due to the impracticality of reconfirming patients, we assessed the positive predictive value by categorizing cases reported as hepatitis B within the KDCA mandatory surveillance system into 2 groups based on whether they were classified as patients. Furthermore, for the cases ultimately considered patients, we reviewed data from medical institutions and individual case reports to ascertain whether they fulfilled the diagnostic and reporting criteria. As such, the positive predictive value of the surveillance system was evaluated.

Data quality is an indicator that reflects the validity and completeness of the information collected in a surveillance system. Its assessment involves evaluating the completeness of data by examining the percentage of missing entries in the system, the thoroughness of the information entered, and similar factors. To evaluate data quality, we calculated the annual registration rate of case reports for patients with hepatitis B collected from the KDCA’s mandatory surveillance system. Furthermore, we assessed the accuracy of the data regarding symptoms through classification into various categories.

Timeliness measures the speed at which a surveillance process is executed, indicating whether the interval between an event’s occurrence and its reporting is appropriate. In the Republic of Korea, medical institutions are required to report cases of hepatitis B to the relevant public health center upon diagnosis, and epidemiological investigations must be initiated within 3 days of patient identification. To evaluate the timeliness of this process, we analyzed the time intervals between the date of diagnosis, the date of reporting by medical institutions, and the date of case report submission to the KDCA mandatory surveillance system.

Usefulness reflects a surveillance system’s achievement of its intended purposes, which may include detecting infectious disease outbreaks and facilitating prevention and intervention programs. To assess usefulness, we examined case reports in which additional precautions and health education measures were implemented.

This evaluation was approved by the Institutional Review Board (IRB) of the KDCA (IRB No: KDCA-2024-05-06-PE-01).

When monitoring under the National Notifiable Disease Surveillance System was initiated in 2000, a sentinel surveillance system was employed to track cases that met the criteria for acute hepatitis B, maternal hepatitis B, and perinatal hepatitis B [11]. In 2011, the surveillance approach for hepatitis B was transitioned to a mandatory system. By 2016, the monitoring of maternal and perinatal hepatitis B was discontinued, with these cases instead managed by the Perinatal Hepatitis B Prevention Program. Currently, only acute hepatitis B cases are under systematic surveillance [8] (Figure 1). Following the reporting of 462 cases in 2011 (the year after the adoption of mandatory surveillance), around 200 cases were reported annually between 2012 and 2015. However, reported cases have increased in recent years, with approximately 400 cases documented annually from 2016 to 2022 [12].

Hepatitis B is included in the Korean notifiable infectious disease reporting system. Therefore, medical institutions must report cases that meet the diagnostic and reporting criteria for this condition to the relevant public health center within 24 hours via the KDCA mandatory surveillance system. The public health center then verifies whether the case is indeed acute hepatitis B, reviews the pathogen report from the diagnostic laboratory, and forwards the case to the provincial government and the KDCA if the criteria are met. Following a final review by the province and the KDCA to confirm adherence to the case definition, the case is then recorded in the official statistics. Cases deemed non-acute are placed in a backlog for re-evaluation by the public health center (Figure 2). Additionally, public health centers are obligated to conduct an epidemiological investigation within 3 days of each reported case. The resulting case report should include a review of the patient’s clinical symptoms, laboratory tests, immunization history, and the suspected cause of infection [8,13].

In the notifiable disease surveillance system of the Republic of Korea, cases of infectious disease are categorized as patients, suspected patients, or pathogen carriers. A case is typically classified as a patient if it meets specific clinical symptom criteria and laboratory test result standards. For the hepatitis B surveillance system, the diagnostic criteria and case classification may appear straightforward, as only confirmed patients must be reported. However, the process is complicated by the diverse characteristics of diagnostic tests for hepatitis B and the required reporting of only acute cases. The diagnostic and reporting criteria for hepatitis B, as outlined in the KDCA guidelines, are detailed in Table 1 [13].

Although these guidelines list jaundice, hematuria, anorexia, nausea, myalgia, severe fatigue, and right upper quadrant tenderness as clinical signs of hepatitis B, the presentation of this condition is often nonspecific compared to other infectious diseases. Consequently, the diagnostic significance of these clinical signs remains ambiguous [14]. Furthermore, it is estimated that about 60% of acute hepatitis B infections are asymptomatic. Current infectious disease surveillance systems mandate the reporting of confirmed cases only, without consistent criteria for the reporting of asymptomatic patients. Notably, asymptomatic status is recognized as a clinical manifestation in the diagnosis and reporting criteria outlined in the guidelines [13]. Hepatitis can cause elevated liver enzyme levels even in the absence of overt symptoms. Due to the nonspecific clinical signs associated with acute hepatitis B, liver enzyme tests are incorporated into the diagnostic and reporting criteria in some countries, such as the United States [15]. As we discuss later, our analysis reveals that asymptomatic infections may be reported as patients within the surveillance system or be excluded from reporting.

For the effective surveillance of acute hepatitis B, it is crucial to identify and exclude patients with chronic hepatitis. To achieve this, positive tests for specific antibodies (immunoglobulin [Ig] M anti-hepatitis B core [HBc]), which are markers of acute hepatitis B, are classified as reportable events [16]. However, frontline healthcare providers frequently use the HBsAg test for chronic hepatitis B if they are unaware that the surveillance system is intended solely for reporting acute hepatitis B. To prevent the inclusion of chronic cases, the guidelines stipulate that case diagnosed with hepatitis B more than 6 months prior should be excluded from the system. Nevertheless, for patients who are not under regular follow-up, accurately confirming their diagnosis can be challenging, as it relies on the patient’s testimony.

The results of multiple diagnostic tests often must be interpreted to accurately identify acute hepatitis B. Various tests are available for hepatitis B, including HBsAg, IgM anti-HBc, HBeAg, and HBV-DNA. The combination of these test results can influence the interpretation of the patient’s condition [15]. Consequently, a positive result for IgM anti-HBc does not necessarily indicate acute hepatitis B. Some countries have incorporated additional tests, including specific antibody assays, into their diagnostic and reporting criteria for acute hepatitis B [17]. Furthermore, the IgM anti-HBc test, commonly used as a marker for acute hepatitis B, can also yield a positive result during the exacerbation of a chronic case [6,14].

To evaluate the positive predictive value within the surveillance system, we determined the proportion of cases that were classified as acute hepatitis B. This percentage was calculated in reference to all hepatitis B reports from medical institutions registered in the KDCA mandatory surveillance system, including cases excluded from the official statistics. For each deleted report, we confirmed the specific reason for exclusion. Overall, 3,142 of the total 5,117 reports registered in the KDCA mandatory surveillance system and public health records were included in the statistics, yielding a positive predictive value of 61.4% within the surveillance system. Among the 1,975 cases reported by medical institutions but excluded from the statistics, reasons for exclusion included chronic hepatitis B (12.4%) and failure to meet the diagnostic criteria (11.9%) (Table 2).

Even within the released statistics, the current surveillance system includes a mix of acute and chronic hepatitis B cases. To remove chronic cases, we excluded reports containing specific keywords such as “carrier,” “chronic,” and “maternal family history” from the data provided by medical institutions and public health centers. We then reviewed the appropriateness of the pathogen reports to reclassify any cases incorrectly identified as acute hepatitis. The analysis revealed that out of 3,142 cases reported as acute hepatitis B, 176 were determined not to be acute. Consequently, the positive predictive value for acute hepatitis B was calculated to be 94.4%, demonstrating that the report data from medical institutions and public health centers had a high positive predictive value (Table 3).

From 2013 to 2022, a total of 3,142 cases were categorized as acute hepatitis B, with 3,097 case reports registered in the KDCA mandatory surveillance system during the same timeframe. Cross-referencing report data from medical institutions with these case reports indicated that case reviews were conducted for 2,802 (88.5%) of all reports included. Further analysis by year revealed a notably high rate of case report registration from 2013 through 2019, with reviews conducted for most cases (>95%). However, from 2020 to 2022, coinciding with the onset of the coronavirus disease 2019 (COVID-19) pandemic, the rates of case report registration were significantly lower than in previous years (Table 3).

Notably, the existing hepatitis B case report form does not provide an option to enter “asymptomatic” in the section for clinical symptoms. Upon review, we categorized the descriptions into alternative symptom categories and determined that 74.0% of the descriptions were accurately classified. Other inputs, including entries indicating asymptomatic status and elevated liver enzymes, were also counted as clinical symptoms (Table 4).

To evaluate timeliness, we analyzed the interval between the diagnosis date at a medical institution and the report date. Our analysis revealed that 91.7% of the 2,880 cases were reported within 1 day of diagnosis, while 8.3% were reported more than 2 days after diagnosis. A year-by-year analysis indicated that the timely reporting rate declined during the COVID-19 pandemic, as depicted in Figure 3. Despite the downturn observed during the pandemic, the overall timeliness remains high. Nonetheless, when compared to the very high timely reporting rates for waterborne and foodborne illnesses cited in previous studies (96.0%), the observed timeliness was inferior [18].

Our analysis of 2,794 registered case reports for timeliness revealed that 2,094 (74.9%) were registered within 3 days of the report date, while 25.1% took more than 4 days to be registered. When examined by year, the rates of timely reporting showed a continuous increase until 2019, prior to the COVID-19 pandemic. However, a decline in timely reporting was observed beginning in 2020 (Figure 3).

Based on data from medical institutions, an analysis of 75 delayed reports out of 262 cases indicated that the most common reason for delay—accounting for 29.3% of instances—was the time required to confirm pathogen reports and specific antibody test results (Table 5).

To evaluate the usefulness of the surveillance system, we extracted keywords such as “guidance,” “education,” “recommendation,” “monitoring,” and “practice” from the final comments provided by the public health center in the case report. We then categorized the cases based on whether additional measures were implemented. Of 2,802 cases, 257 (or 9.2%) had documented actions taken. More specifically, 7.9% involved the provision of health education, which covered topics such as preventive measures, personal hygiene, precautions to avoid blood and body fluid exposure during sexual contact, and the importance of regular medical appointments. Additionally, 3.0% reported recommending vaccination to individuals who were unvaccinated, as well as the provision of further tests and vaccinations for family members or close contacts. In some cases, both health education and vaccination guidance were provided.

This evaluation used data from the KDCA’s mandatory surveillance system to review the monitoring framework for acute hepatitis B. The analysis indicated that the positive predictive value, data quality, and timeliness were generally satisfactory. However, potential areas for improvement were recognized, particularly in terms of simplicity and usefulness.

We examined the diagnostic reporting criteria to assess their simplicity and to identify potential improvements. In the current surveillance system for acute hepatitis B, cases are reported based solely on clinical symptoms and diagnostic findings. However, these clinical symptoms are often nonspecific; furthermore, diagnostic testing methods can vary, and confirmation of past infection history is necessary. These features contribute to a risk of misreporting, as patients with chronic hepatitis B—who account for most hepatitis B cases—may be incorrectly reported as having the acute disease. This conclusion is supported by our evaluation of positive predictive value, which indicates that many chronic cases are reported, although they may not ultimately be misclassified as acute. Additionally, we noted some confusion within the surveillance system due to unclear reporting criteria for asymptomatic patients.

The evaluation of positive predictive value did not confirm whether cases were acute; rather, it focused on instances of misreporting within the surveillance system. The analysis revealed that an estimated 94.4% of cases in the official statistics were acute hepatitis B, and the majority were correctly triaged by the surveillance system. However, 38.6% of cases initially reported by medical institutions represented misreporting and were excluded from the statistics. The issues identified in the diagnostic and reporting criteria introduce confusion between acute and chronic cases, potentially decreasing the positive predictive value of the surveillance system. This could lead to overestimation due to the inaccurate classification of cases as patients within the system.

Data quality was evaluated by analyzing the case report registration rate and reviewing the categorization of symptom inputs. Our comparison of the case report registration rate with data from medical institutions revealed a high rate of 88.5%. Notably, nearly all cases (over 99%) were registered prior to the COVID-19 pandemic, whereas only approximately 72% were registered after the pandemic’s onset. This suggests that case studies may not have been conducted or registered due to operational changes necessitated by the urgent actions of public health centers during the pandemic. Additionally, we identified areas for improvement in the case report entry process. For instance, some cases of asymptomatic hepatitis B were reported, but these were not appropriately classified within the system’s symptom-related data.

In the analysis of timeliness, the timely reporting rate from medical institutions was 91.7% and the case report timeliness was 74.9%, indicating a generally high level of promptness. However, timeliness has declined at both medical institutions and public health centers since the onset of the COVID-19 pandemic, highlighting the need for measures to improve the rate of timely reporting. Nevertheless, the surveillance system’s timeliness is considered excellent, as over half of reported cases are investigated within 24 hours. With the initiation of the national vaccination program, the HBsAg positivity rates have continuously declined, suggesting a potential reduction in the public health risk associated with hepatitis B. Therefore, further discussion is warranted regarding the significance of prompt reporting of hepatitis B.

In evaluating the usefulness of the surveillance system, incorporation of additional measures such as health education and vaccination guidance was found to be infrequent. Due to the effectiveness of vaccination against hepatitis B and its transmission primarily through blood, contact identification is not part of the case report form, and investigations of risk factors are cursory. Therefore, the original purpose of the acute hepatitis B surveillance system reflects simple outbreak surveillance, and the system is likely underutilized.

We did not assess the representativeness of the surveillance system, which is the degree to which the collected information accurately reflects the disease. Chronic cases of hepatitis B are generally prioritized for management. However, the surveillance system in the Republic of Korea is limited to monitoring acute cases, potentially yielding an incomplete representation of hepatitis B in the country. The WHO has set a goal for the global elimination of viral hepatitis, for which it has identified 4 key indicators: vaccination, prevention of perinatal infection, diagnosis, and treatment [3]. However, surveillance systems that focus solely on acute hepatitis B fail to capture these indicators. Therefore, the current acute hepatitis B surveillance system in the Republic of Korea does not encompass the management of all aspects of hepatitis B, nor does it contribute effectively to the prevention and management of chronic hepatitis B. Direct management of patients with chronic hepatitis B, who are estimated to make up about 2% of the country’s total population, presents a challenge. Nonetheless, the current system requires improvements to better support the national strategy for viral hepatitis management.

Notably, the WHO guidelines for eliminating viral hepatitis also emphasize the importance of acute viral hepatitis surveillance [3]. Furthermore, in countries with low HBsAg positivity rates and established vaccination programs, horizontal transmission is generally the predominant route of hepatitis B transmission. Therefore, monitoring horizontal transmission in the Republic of Korea and determining the extent of acute hepatitis B in the community is essential [4]. Improving the positive predictive value and sensitivity of the current surveillance system would provide a more accurate evaluation of the situation and yield meaningful insights.

Based on the findings of this evaluation, we propose the following improvements. (1) The current surveillance system for acute hepatitis B does not appear to adequately capture the epidemiological characteristics of acute cases. To accurately reflect the true incidence and epidemiological profile of hepatitis B, which should be the primary goal of surveillance, improvements are necessary. This includes the incorporation of additional liver function parameters such as aspartate aminotransferase, alanine aminotransferase, and bilirubin levels—parameters that are already monitored in the surveillance of other acute viral hepatitis infections like hepatitis A and E. While the inclusion of liver function levels in diagnostic reporting may complicate the criteria, we contend that it will increase the system’s accuracy and positive predictive value. Such improvements could also refine the current system, which lacks clear reporting criteria for patients with asymptomatic acute hepatitis B who exhibit no clinical symptoms but present with elevated liver enzyme levels. (2) The acute hepatitis B surveillance system should be improved to promote the elimination of viral hepatitis and the management of chronic hepatitis B. While acute hepatitis B can lead to acute liver failure in some cases, it does not have the long-term outcomes of the chronic type, which poses a much greater public health challenge due to its prolonged disease burden. Methods must be developed that utilize existing surveillance systems for the more effective management of viral hepatitis. For instance, incorporating mechanisms to monitor the progression from acute to chronic hepatitis B would increase the usefulness and flexibility of the surveillance system. (3) An integrated viral hepatitis surveillance system may be necessary in the Republic of Korea. At present, only acute cases of hepatitis B are monitored, which inevitably leads to the misreporting of chronic cases. This issue could be mitigated by further educating stakeholders. However, most developed countries also encounter this challenge in their surveillance efforts for acute hepatitis B. The nonspecific nature of symptoms and the lack of a clear diagnostic basis make it difficult to definitively correct the misreporting of chronic cases [14,19]. However, the exclusive surveillance of acute hepatitis B is uncommon among industrialized nations, many of which have established complex surveillance systems that encompass perinatal, maternal, and chronic hepatitis B [20]. Until 2015, the Republic of Korea also conducted surveillance of perinatal and maternal hepatitis. Although these surveillance activities have been discontinued, cases of perinatal and maternal hepatitis B continue to be managed through the perinatal infection prevention program. For comprehensive hepatitis B management, particularly in a nation where the prevalence of chronic cases is at least 2%, it is crucial to monitor chronic infections. Individuals with chronic hepatitis B require regular checkups to prevent liver disease and to facilitate early detection of HCC. In the Republic of Korea, data on hepatitis B treatment are available from the Health Insurance Review and Assessment Service. Therefore, an integrated viral hepatitis surveillance system that consolidates the existing fragmented surveillance and management frameworks would enable more efficient national management of viral hepatitis. (4) Evaluations of data quality and timeliness indicate a decline in surveillance performance since the onset of the COVID-19 pandemic. This issue is believed to have affected not only hepatitis B surveillance systems but also those developed for other infectious diseases, due to disruptions stemming from the COVID-19 response. All parties involved in disease surveillance, including medical institutions and public health centers, must persist in their efforts to return these systems to pre-pandemic standards. (5) We recommend an ongoing evaluation of surveillance systems. The public health surveillance systems are generally evaluated to ascertain their effectiveness and fulfillment of their objectives, to account for shifts in the natural history of the disease under surveillance, or to implement changes to the system [21]. The Korean hepatitis B surveillance system has evolved from a sentinel monitoring approach to a mandatory reporting system. Concurrently, the focus of surveillance has shifted from encompassing perinatal, maternal, and acute cases to exclusively monitoring acute cases. Externally, the system has also been impacted by the COVID-19 pandemic. Despite more frequent changes to the hepatitis B surveillance system than to other infectious disease systems, no evaluation has been conducted since 2009. It is essential to evaluate the surveillance system to understand the implications of these changes on the monitoring of infectious diseases. (6) Although acute hepatitis B may not necessitate an urgent response due to its less severe characteristics, the disease in general remains a significant health concern due to the substantial burden of long-term illness. It may be time to establish renewed, long-term goals and strategies for the control of viral hepatitis on a national scale. Action plans should be developed and implemented with dedicated budgetary allocations, human resources, and other commitments, including health communication and the education of medical professionals.

Nevertheless, our evaluation has several limitations. First, although the CDC guidelines recognize stakeholder involvement as a key component of surveillance system evaluation, this analysis relied solely on data from medical institutions and case reports. Incorporating a survey of health facilities active in the field, public health centers, provincial infection control officers, and medical institutions tasked with case reporting could have provided a more multidisciplinary perspective on the various metrics of the surveillance system.

Second, a calculated positive predictive value can differ from the actual value because the analysis of this metric is based on the review of existing data, rather than on the testing or analysis of patients. The positive predictive value of the surveillance system under evaluation can serve as an indicator of the system’s characteristics. However, if the modeling accounted for the false positive rates of the diagnostic methods, or if a subset of actual patients were selected and retested to identify false positives, a more accurate positive predictive value could be determined.

Third, the current acute hepatitis B surveillance system also includes monitoring of deaths due to acute hepatitis B, yet no separate analysis or evaluation of this aspect has been conducted. Nonetheless, the primary purpose of acute hepatitis B surveillance is simply to understand the scale of its occurrence. We believe that evaluating the performance of the surveillance system in capturing fatal cases of acute hepatitis B may be unnecessary, given that no deaths were reported during the period analyzed in this evaluation.

The current hepatitis B surveillance system in the Republic of Korea is designed to monitor new cases of acute hepatitis B resulting from horizontal transmission. While this surveillance has been effective in reducing the burden of hepatitis B, improvements to the system are still possible. Our analysis unveiled high positive predictive value (94.4%), data quality, and timeliness. However, the objectives of surveillance are not fully achieved due to the potential misreporting of chronic hepatitis B cases, and the usefulness could also be improved. More accurate triaging of acute cases is essential to increase this system’s efficacy as a tool for acute hepatitis B surveillance.

Ethics Approval

This evaluation was approved by the IRB of the KDCA (IRB No: KDCA-2024-05-06-PE-01).

Conflicts of Interest

The authors have no conflicts of interest to declare.

Funding

None.

Availability of Data

The datasets are not publicly available but can be obtained from the corresponding author upon reasonable request.

Authors’ Contributions

Conceptualization: all authors; Data curation: JC; Formal analysis: JC; Investigation: JC; Methodology: JC, MC; Project administration: all authors; Resources: JC, JY; Software: JC, MC; Supervision: JY, ST; Validation: ST; Visualization: JC, ST; Writing–original draft: JC; Writing–review & editing: all authors. All authors read and approved the final manuscript.