Health Resource Management for COVID-19 Using Big Data Model

Abstract

The aim of this study was to apply a big data model for estimating resource requirement for the health sector to deal with the epidemic of coronavirus disease 2019 (COVID-19) in Thailand, with an expectation that the results of this study would lead to the recommendations for the preparedness of adequate health resources for medical and health personnel. It was conducted by using a mixed methods design. The population for the qualitative study were the experts from various technical departments of Ministry of Public Health; and those for the quantitative study were health care facilities from both public and private sectors. The resources specified in this study included (1) particulate respirators (N95), (2) personal protective equipment (PPE), (3) favipiravir, an antiviral drug, (4) ventilators, and (5) beds for patients in cohort wards, isolation rooms and airborne infection isolation rooms (AIIR). Primary and secondary data were collected through a newly developed databased template; and data analysis was performed by using Microsoft Excel and STATA solfwares. The study period was 4 months between January and May 2020, separated into 2 periods, one was the high level outbreak (reproduction number - Rt=1.8), 5-26 April 2020 or weeks 14-17; and the other was the lower-level outbreak (Rt=0.56), 11 May to 7 June 2020 or weeks 20-23. The study found that during the high level outbreak, the needs for N95 and PPE were 58,397, 81,735, 116,191, and 167,710 sets in the week 14 to 17, respectively; and the stock would be inadequate in the weeks 16 and 17. It required additional procurement to fill the gaps. However, the estimated requirement of favipiravir, ventilators, cohort ward beds, isolation rooms and airborne infection isolation rooms was found to be sufficient. It was also found that the estimated resource requirement during the low-level outbreak period (Rt=0.56), did not exceed the availability of the resources (N95, PPE, favipiravir, ventilator, and the 3 types of beds). This study has demonstrated the necessity for close-monitoring of the resources needed for the preparedness and effective response to the epidemic. Therefore, a national COVID-19 data center should be established in order to compile all essential data i.e. a number of confirmed cases and resources availability which could be displayed on real-time basis and disaggregated by province. Moreover, there should be an expert team for analyzing and forecasting numbers of patient, including health resources requirement at both central and provincial levels. This would support the effective preparedness and response of health systems to the outbreak.