Welcome to CompARE!

This tutorial introduces you to CompARE, a web-tool for designing trials with composite endpoints.

Go to CompARE!

What is a Composite Endpoint?

There are some diseases for which more than one clinical outcome is important and where all outcomes are expected to be affected by the treatment. When a single outcome is not sufficient to capture the treatment benefits, the use of more than one primary variable may become necessary1.

Instead of using each as a separate primary endpoint (creating multiplicity) or selecting just one to be the primary endpoint, it may be appropriate to combine those clinical outcomes into a single variable2.

The so-called Composite Endpoint (or composite outcome) consists of two or more component outcomes. Patients who have experienced any one of the events specified by the components are considered to have experienced the composite outcome.

Benefits and Risks

Composite endpoints are frequently chosen as primary endpoint in many health areas and specially in cardiovascular and oncology trials. There are three key advantages for using a composite. First, they avoid the need of multiplicity adjustment. Second, a composite endpoint contains more information on the course of the disease than a single endpoint providing a better explanation about the differences between treatment groups. Third, the increment of the number of observed events is expected to increase the power.

The main drawback of using a composite endpoint is its interpretation since its components rarely have comparable clinical importance and similar treatment effects. The addition of components that are not relevant enough could compromise the interpretation of results and reduce power. Hence, the choice of the particular components for the composite has a great importance in the design phase.

Why use CompARE?

CompARE is intended to provide guidance on how to deal with composite endpoints in the planning and statistical analysis of clinical trials. CompARE helps researchers to study:

What do we need to use CompARE?

Structure of CompARE

CompARE is composed of three main parts:

  • Input panel
  • Menu bar
  • Output panel


The user is prompted for the anticipation of different parameter values summarized in the following section.

Input Panel

For each Composite Component, you need:

  • To provide the point value or interval of values for the probability of the Endpoint.
  • To anticipate the point value (or interval of values).
  • To provide which effect measure you will use (risk ratio, risk difference or odds ratio).
  • To anticipate the expected effect.


For the Composite Endpoint, you need:

  • To provide which effect measure you will use (risk ratio, risk difference or odds ratio).


Additionally, you need to provide the expected Association between the Composite Components in terms of Pearson’s correlation.


For calculating the sample size, you need:

  • To provide the significance level.
  • To give the desired power
  • To provide which estimate for the variace you will use (pooled or unpooled variance estimate)


Learn more in:

Input panel in CompARE.

What does CompARE do?

Key messages and short report

In the first tab, CompARE shows a key messages about the trial design as well as a short report according to the parameters anticipated in the Input Panel.

Association between components

When using a composite binary endpoint, one needs to take into account the degree of association between the composite components. However, this association is usually unknown or difficult to anticipate.

By means of CompARE, you can calculate which values the association between the components could have based on the marginal parameters. The association is specified using Pearson's correlation measure and Relative Overlap.

Learn more in:

Association measures for Binary Endpoints using CompARE.

Sample Size

When designing a study, investigators need to determine how many subjects should be included. By enrolling too few subjects, a study may not have enough statistical power to detect a treatment difference. Enrolling too many patients can be unnecessarily costly or time-consuming.

To size a trial with a composite binary endpoint, one needs to specify the event rates and the effect sizes of the composite components along with the correlation between them. In practice, the marginal parameters of the components can be obtained from previous studies or pilot trials, however, the correlation is often not previously reported and thus usually unknown.

By means of CompARE, the user can calculate the required sample size for composite binary endpoints based on the anticipated information on the components in cases where the correlation is totally or partially known, as well as where there is uncertainty in the event rate values of the components.

Learn more in:

Sample size calculation for Composite Binary Endpoints using CompARE.

Effect Size

Before a study is conducted, investigators need to anticipate which is the minimum effect relevant to be detected. Depending on this effect, analysis and sample size will be carried out.

By means of CompARE, you can calculate the effect size for a composite binary endpoint based on the marginal parameters of its components.

Learn more in:

Effect size calculation for Composite Binary Endpoints using CompARE.

Endpoint Selection

CompARE can help you to make a more informed decision on the Primary Endpoint you should use in your study.

Considering the Endpoint 1 as the indispensable endpoint you must use in your trial to test the treatment efficacy, CompARE will help you to choose between a Endpoint 1 and the Composite Endpoint as the primary endpoint for the trial.

Learn more in:

Endpoint Selection in trials involving Composite Binary Endpoints using CompARE.

Glossary of Terms


  1. Guideline on multiplicity issues in clinical trials. European Medicines Agency, 2017. (link)

  2. Multiple Endpoints in Clinical Trials. Guidance for Industry. Food and Drug Administration, 2017. (link)