A **survival curve** is a chart that shows the proportion of a population that is still alive after a given age, or at a given time after contracting some type of disease.

This tutorial shows how to create a survival curve in Excel.

**Creating a Survival Curve in Excel**

Suppose we have the following dataset that shows how long a patient was in a medical trial (*column A*) and whether or not the patient was still alive at the end of the trial (*column B*).

In order to create a survival curve for this data, we need to first get the data in the correct format, then use the built-in Excel charts to create the curve.

**Formatting the Data**

Use the following steps to get the data in the correct format.

**Step 1: List all of the unique “Years in trial” values in column A in column D:**

*Note: Always include “0” as the first value.*

**Step 2: Create the values in columns E through H using the formulas shown below.**

*Note: we removed the value “18” in column D since there were no death values associated with that time.*

Here are the formulas used in the following cells:**E3:** =COUNTIFS($A$2:$A$16,D3,$B$2:$B$16,1)**F2: **=COUNTIF($A$2:$A$16, “>”&D2-1)**G3:** =1-(E3/F3)**H2:** =1**H3:** =H2*G3

In order to fill in all of the other values in column E, simply highlight the range E4:E13 and press Ctrl-D. Fill in all of the other values in columns F through H using the same trick.

Now we’re ready to create the survival curve.

**Creating the Survival Curve**

Use the following steps to create the survival curve.

**Step 1: Copy the values in columns D and H into the columns J and K.**

**Step 2: Copy the values in the range J3:J13 to J14:J24. Then copy the values in the range K2:K12 to K14:K24.**

**Step 3: Create a list of values in column L as shown below, then sort from smallest to largest values in column L:**

**Step 4: Highlight cells J2:K24, then select “Insert” > “Charts|Scatter” > “Scatter with Straight Lines and Markers” option.**

Feel free to modify the title, axes names, and chart colors. The resulting chart will look something like this:

Informative and a quick way to provide a very rough survival curve. Note that your formula in Step 3 is susceptible to an error. H3: =H2*G3 clearly outputs 0 if 1-(died/alive) is 0. This means that if died = alive (ie the same amount of subjects died as the same number alive at that time period), then the survival curve drops to 0 thereafter. Good job though, instructions are easy to follow!