This tutorial will show you how to set up a cumulative incidence analysis in Excel using the XLSTAT add-on statistical software.
Dataset for running a Cumulative Incidence analyses
An Excel sheet with both the data and results can be downloaded by clicking here.
The data have been obtained from Marubini and Valsecchi (1996) and represent a randomized clinical trial investigating the effect of a treatment on the time of appearance of a set of symptoms on patients. Two groups of patients are studied, the first being treated and the second being a control group.
The cumulative incidence applies in cases where you are in the presence of competing events (known as competing risks), that is to say, when several events can occur in addition to censorship.
In our case, it is either a local relapse (event 1), or the appearance of new metastases (event 2) or a censoring (cure or lost sight of by those responsible for the study).
Setting up a Cumulative Incidence analyses
After opening XLSTAT, select the XLSTAT / Survival analysis / Cumulative Incidence command.
Once you've clicked on the button, the Cumulative Incidence box will appear.
Select the data on the Excel sheet. The Time data corresponds to the durations when the patients either relapsed locally (event 1),or have new metastases (event 2) or were censored (event 0).
The "Status indicator" describes whether a patient locally relapse (event code=1), has new metastasis (event code=2) or was censored (censored code = 0) at a given time.
So that XLSTAT takes into account the information whether the patient belongs to the control or the treated group, we need to select the groups information.
The computations begin once you have clicked on OK. The results will then be displayed on a new Excel sheet.
Interpreting the results of a Cumulative Incidence analysis
The results for the first group are displayed first (treatment). The first table displays a summary of the data for the treated patients.
The next table corresponds to the "Cumulative incidence table" for event 1 in treated group. It contains the results of the cumulative incidence with several key indicators.
The next table corresponds to the cumulative survival function for event 1 in the treated group.
Then, we can visualize the cumulative incidence function and the cumulative survival function, bounded by the confidence intervals. The circles identify the censored data.
Next, the same series of results is displayed for the second event for the treated group.
Then, we can compare the two events for the treated group. We can see that new metastases have a greater incidence than local relapses.
Next, the same series of results are displayed for each of the events for the control group. The comparison plot shows that both groups have the same behavior with respect to each event.
Last the comparison of the two groups for each event is of great interests. On the following curves, we can see that for both events (local relapse and new metastases), the treatment impacts significantly negatively the incidence of the events and impacts significantly positively the survival time of the patients.