nci logo
NIH
U.S. National Institutes of Health National Cancer Institute

SEER*Stat Trends Exercise 1b

Problem Statement

In exercise 1a we calculated trends for malignant colon cancer for males in the SEER 9 registries for the entire time period 1975-2006. By reviewing the rates in exercise 1a, it is clear that there are two overall distinct trends over those years. The rates are generally rising from 1975 until the mid-1980s (peaking in 1985), then they begin declining for the remaining years. In this exercise, we will use SEER*Stat to calculate trends for two time periods, 1975-1985 and 1986-2006.

Is the APC for 1986-2006 significantly different from the APC for 1975-1985 (p < 0.05)?

Key Points and Reminders

Use the dictionary editor to create a variable called "Year of diagnosis (75-85, 86-06)". This variable should have 2 groupings, one for each of the time periods. Once this variable is created, it will appear as a choice in the Trend Variable control on the Statistic Tab.

SEER*Stat can test whether an APC over one time period significantly differs from an APC for a non-overlapping time period. The first time period defined in the trend variable is used as the base time period and each subsequent time period is compared to the base.

SEER*Stat cannot determine when the changes in trends occur. This is one of the main differences in SEER*Stat and the Joinpoint Trend Analysis Software (http://srab.cancer.gov/joinpoint/). Therefore, for this exercise we have pre-determined the two time periods.

SEER*Stat's export feature provides a method to create a generic ASCII text file containing the age-adjusted rates. This file can be used as the input data file in Joinpoint.

Step 1: Open Exercise 1a's Matrix and Retrieve the Session

SEER*Stat matrix files include the session information used to generate the table. This information serves as documentation for the results and provides a convenient method for generating similar statistics.

  • Open your saved matrix file from exercise 1a. If you did not save the file, use the file key.trends1a.sim.
  • Open the Matrix menu and choose Retrieve Session.
  • Two windows should now be open in SEER*Stat. Close the Matrix window and begin working in the Rate Session window.

Step 2: Dictionary

  • Use the Dictionary icon, or select Dictionary from the File menu, to open the Dictionary.
  • Open the "Race, Sex, Year Dx, Registry, County" folder, click on the "Year of Diagnosis" variable, then click Create.
  • Name the new variable "Year of diagnosis (75-85, 86-06)".
  • Delete all of the existing groupings.
  • Highlight all of the values from 1975 to 1985 (inclusive) and click Add. Title the new grouping "1975-1986".
    • To highlight a continuous range of values, click on the first value in the range, then hold down your SHIFT key while you click on the last value in the range.
  • Highlight all of the values from 1986 to 2006 (inclusive) and click Add. Title the new grouping "1986-2006".
  • Click OK to save the user-defined variable, then click Close to exit the Dictionary.

Step 3: Statistic Tab

On the Statistic tab:

  • Change the Trend Variable to the "Year of Diagnosis (75-85, 86-06)" variable that you just created.

Step 4: Output Tab

On the Output tab:

  • Edit the title of your matrix, if desired.

Step 5: Execute the Session

  • Use the Execute icon, or select Execute from the Session menu, to execute the session.

A dialog will display the progress of the job. When the job completes, a new window will open containing the output matrix.

Step 6: Check Your Results