How to compare groups of patients?

Modified on Wed, 4 Oct, 2023 at 8:17 AM

You can compare groups of patients on several variables quickly with the "Compare groups" feature.


Here is how it works:

  1. Head to the menu "Statistics" and then "Compare groups"
  2. Choose the variable that will create your groups.
    For example, if you want to compare men and women, choose the variable "Gender"
  3. Choose the variables you want to compare between the groups
  4. Click on "Compare groups"


Frequently asked questions


Can I compare more than 2 groups?


Yes, you can compare more than 2 groups.

In step 2, when you choose the variable to create groups, you can pick a "List" variable with more than 2 modalities.

Or you can create more than 2 groups from a Numeric variable by "cutting" your variable into several groups: How to create groups from a Numeric variable? (Numeric Variable Discretization).


Can I create groups from a Numeric variable?


Yes, you can create groups by using a Numeric variable and defining thresholds.

In step 2, choose a numeric variable and follow the instructions. More information can be found here: How to create groups from a Numeric variable? (Numeric Variable Discretization)


Can I merge groups together?


Yes, you can reorder the List variable that has been used to create your groups: How to group values (modalities) of a List variable?


Can I have charts and "what to write" texts?


Yes. Next to each p-value, you can find an icon that allows you to expand the result and see more details: charts, what to write in methods, what to write in results, ...


What is "Group comparability"?


In addition to the variables you want to compare, the application automatically compares your groups on demographic data like age at inclusion, follow-up duration, ...

It is important that your groups are comparable before comparing outcomes. Else, you may have confounding variables that will affect your outcomes.


Is it possible to perform a case-control study or a propensity score matching?


Yes, you can use propensity score matching to match groups of patients.

This only works with 2-group comparison.

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article