According to Baseline 101 – who is who? by Nadine Marlin and John Allotey, baseline tables show the characteristics of research subjects included in a study. Depending on the study design they may have specific purposes but generally they show if the population included conforms to the eligibility criteria of the study. They also indicate what population the results could be generalised to. Baseline characteristics table are very important very in large cohort studies and clinical trials. You can read more about requirements of baseline characteristics here.
- Upload your .csv file which should typically consists of subject IDs, grouping variable (categorical), and other variables (categorical or continuous). The sample data consists of subject IDs, grouping variable (BMI_Group), and other variables (Age, Height, Weight, BMI).
-
Now, we have to choose:
a. Categorical grouping variable: BMI_Group
b. Categorical Variables: Gender
c. Excluding Variables: Sub_ID
Here, specify your grouping column, categorical variable columns and columns to be excluded. We need not specify the continuous variables.
Here is the output data. Click here
The tool was developed for internal use within the SANSCOG study. I would like to thank Suhrud Panchwagh for conceptualizing the idea and writing the backend R code for conducting statistical analysis and generating flextables. My contribution to the project was optimizing the backend code for flexible analysis and deploying the project on RShiny.