One of the most important to be considered during taking care of COVID-19 patient is to understand which patients require more intensive treatment and attention. Researchers in Oklahoma State University’s center for health system innovation ( CHSI) is performing research by applying big data analytics to develop predictive models of COVID-19 patient risk which can help physicians manage inpatient in a better way during the pandemic.
Zhuqi Mio who is the health data science program manager at CHSI and Meghan Sealey a doctoral student studying statistics at Oklahoma state has worked with anonymized data which was about 19,000 COVID-19 patients datasets from the healthcare IT firm Cerner’s. with the help of these data they developed two tools for modeling mortality risk, one is based on the patient data during the time of admission and the other based on patient data from first data during hospitalization.
These models identified a similar set of medical conditions which was suggested by the center for disease and prevention such as the essential risk factors for the death, history of diabetes, respiratory issues, hypertension, and also kidney failures, and the researchers also found some unique ones.
The researchers said that by using these tools they were able to accurately predict mortality nearly 70% of the patients for the first model and the same 70% for the second data. The team found a turn of benefits in store for the healthcare industry if it was deployed these tools in the field.
William Paiva CHSI’s executive director said that these analytic tools are the wave of the future for diagnosing, staging, and monitoring disease progression and save the lives of patients and it also helps to reduce the financial burdens for both the patients as well as for the healthcare system.
Miao added that there is always an urgent need to determine which COVID 19 patients are suffering and have the highest risk for bad clinical outcomes as early as possible so that it can help in planning and perform an action to save more lives. Since there rise in COVID cases in the US as reported so there is no time better than now to develop a tool that can predict which patients are at most risk.
