An evaluation of coronavirus disease 2019 (COVID-19) surveillance system, Nopparat Ratchathani Hospital, Thailand, 10 January–30 April 2020

Abstract

Background: Coronavirus disease (COVID-19) is an emerging infectious disease that has a worldwide impact. After China’s report, the Department of Disease Control (DDC) established surveillance system at ports of entry and all hospitals in the country in January 2020. Screening of suspected COVID-19 patients in hospitals and rapid laboratory confirmation is critical for preventing wide disease spreading. Objectives of this study were to understand process of the surveillance system, evaluate quantitative attributes of suspected COVID-19 case reporting and qualitative attributes of the system at Nopparat Ratchathani Hospital, and provide recommendations for system improvement.
Methods: Quantitative Cross-sectional and qualitative studies were conducted. The definition of a suspected COVID-19 case used for evaluation was a patient visiting the hospital during January–April 2020 with symptoms and history meeting the DDC’s case definition of “Patient under investigation”. Quantitative analysis were used to assess sensitivity, predictive value positive, data quality, and timeliness. To evaluate qualitative attributes including usefulness, simplicity, flexibility, acceptability and stability, we interviewed personnel involved in the surveillance system and conducted thematic analysis.
Results: The surveillance system in this hospital had screening process for suspected COVID-19 patients and specimen collection for laboratory confirmation. It was a passive surveillance system with a reporting sensitivity of 88.71% and a positive value predictive of 89.08%. Unreported suspected cases were among the severe and fatal pneumonia cases. Variables, including gender, nationality, occupation, address (province), admission date and date of onset, had 100% completeness while 91.38% was for age variable. The accuracy of variables including age, admission date, and onset date were 94.34%, 93.40%, and 78.30%, respectively. For the timeliness, 88.88% of cases were reported to the national database on the day of hospital admission. Gender, age, province and week of onset variables had well representativeness. Qualitative findings suggested that the process of reporting suspected cases had complicated workflow and multiple steps. Electronic mail reporting were used.
Conclusion: Increasing surveillance among hospitalized pneumonia patients, developing full-online database and reporting system of COVID-19 surveillance, and combining laboratory and epidemiological databases are keys recommendations to make the surveillance system more sensitive and effective.