Background
There is a limited amount of studies that focus in-depth on if there are disparities in gender and ethnicity when analyzing the relationship between common key risk factors such as diabetes, hypertension, and obesity and COVID-19 positivity.
We find this to be important because medical insight should be holistic and take into account people of various health conditions, severities, and backgrounds, especially when it pertains to a disease that is so widespread.
-
There are millions of COVID-19 test results and EMR data that is not used for critical analysis in determining common key risk factors for COVID; very little research is done on finding common key risk factors pertaining to the patients’ backgrounds
-
Most research show common key risk factors strictly in the context of severe COVID/mortality risks of people with pre-existing conditions and exclude mild cases of COVID
Inclusion and Exclusion Criteria
In order to determine if the prevalence of hypertension, diabetes, and obesity are different between COVID-Postive and COVID-Negative patients, we included any patient in the data set who has taken a COVID test and had their data recorded. Patients were defined with these things in mind:
-
Had a unique id number
-
Were tested for COVID
-
Returned a positive or negative COVID test result
Any data that is incomplete will be excluded in our data.
Patients with severe COVID is defined as those with the endpoints:
remdesivir administration, extracorporeal membrane oxygenation (ECMO), mechanical ventilation, COVID pneumonia ARDS, and death
-
To collect the data we needed, we wrote a query in Microsoft SQL that gathered all of the patients in the UC CORDS database who were tested for COVID-19 and organized them into a data set based on their IDs, Gender, Race, COVID positivity status, and ICD10 codes for our risk factors/severe COVID outcomes.
-
The risk factors that we used were : Diabetes (Type 1 and 2), Obesity and Hypertension.
-
The Severe COVID outcomes we used were Acute Respiratory Distress Syndrome (ARDS), and Dependence on Ventilators.
-
We imported this dataset into Statistical Analysis Systems (SAS) to perform cross tabulations (PROC FREQ) with Chi Square analysis.
-
We mainly looked at relationships between COVID positivity and our risk factors stratified by gender and race. We furthered this by looking at the relationship between severe COVID outcomes (ARDS, Ventilators) and our risk factors among patients who tested positive.
Statistical Analysis Plan
Raw Analysis
