How to create a mean cumulative functions figure?

[SeaMoon_168]

I aim to create a mean cumulative functions figure shown below. Now, I can create a CIF figure. Could anyone help me figure out how to create it? Many thanks!

[PaigeMiller]

Please show us a portion of the SAS data set involved. Please follow these examples and instructions to provide a SAS data set in a form we can use.

[SeaMoon_168]

Thank you for your reply. Actually, I have no clue what kind of data required to create this figure. As I mentioned, I only know how to create the Cumulative Incidence Function (CIF) figures. The raw data are regular survival data including event (0, 1), treatment groups, and follow-up time. 

[PaigeMiller]

Is not the plot based upon the CIF figures?

[SeaMoon_168]

I guess so though I am not 100% sure. The Y-axis of the CIF plot is generally 0.0 - 1.0 as the marginal probability. However, Y-axis of this plot is 0-100 as events per 100 patients. I am not sure if multipling the CIF marginal probability (0-1) could be the Y (0-100) of this figure? That confused me.

[Quentin]

Are you asking how to create the graph, or are you asking how to interpret the graph, or are you asking how to calculate the statistics that would be needed to create a graph like that?  What data are you starting with?

Is this a homework problem for a class, or a work problem where you've been asked to generate a graph that you don't fully understand?  Either way, I would refer to the instructions that came with the request, or talk through your understanding of the graph with the requestor.

[Quentin]

---

@SeaMoon_168 wrote:

I guess so though I am not 100% sure. The Y-axis of the CIF plot is generally 0.0 - 1.0 as the marginal probability. However, Y-axis of this plot is 0-100 as events per 100 patients. I am not sure if multipling the CIF marginal probability (0-1) could be the Y (0-100) of this figure? That confused me.

---

That sounds likely.  You're used to graphs that display the probability as a proportion (0.0-1.0) but this graph shows the probability as a percentage (0-100).  Should be fine either way.  Sometimes for rare events, rates will be displayed as number of patients per 1000, or 100,000,  or whatever the author thinks is convenient.

[SeaMoon_168]

Yes, I am not sure if the figure listed in my original post is equivalent to the CIF figure. I saw this kind of figures per 100 patients listed in a couple of journal papers. So it could be popular. if no, how to create this kind of figure in SAS?

[Reeza]

Looks like a failure plot (opposite of survival plot) with the HR added in. This would definitely be a custom graphic created in SAS but could be done.