National Cancer Institute Home at the National Institutes of Health | www.cancer.gov

DevCan Frequently Asked Questions

  1. Why don't the numbers generated in DevCan match the numbers provided in the SEER Cancer Statistics Review?
  2. Why are there separate databases with Kaposi Sarcoma and Mesothelioma classified as individual cancer sites? What are the effects of having separate databases with these sites broken out?
  3. Why don't the numbers generated by the current version of DevCan match those generated by previous versions using the same data?
  4. How has data provided with DevCan 6.x changed from earlier versions?
  5. What does the follow-back year define?
  6. Why won't the data I used in DevCan versions 4.x work with version 5.1 or later?
  7. How do I import data from SEER*Stat into DevCan?
  8. Is there a suggested citation for DevCan?

1. Why don't the numbers generated in DevCan match the numbers provided in the SEER Cancer Statistics Review?

In most cases, this occurs because the database or selections do not match the ones used in the SEER Cancer Statistics Review (CSR)External Web Site Policy. Make sure that the database selected in DevCan is the exact database referenced in the CSR. Also, check that the year, race, sex, and age combinations you are using in DevCan matches those in the CSR. If viewing lifetime risk estimates, be sure that the check box labeled"Last Interval Open Ended" in DevCan's Options dialog is checked. In the CSR, the risk of developing cancer comes from SEER areas only, while the risk of dying of cancer comes from the total US.

[Return to top]

2. Why are there separate databases with Kaposi Sarcoma and Mesothelioma classified as individual cancer sites? What are the effects of having separate databases with these sites broken out?

Deaths caused by Kaposi Sarcoma and Mesothelioma were not classified as such until 2000, so the only data available for these sites is since that year. To provide risk estimates from earlier years, databases were provided with DevCan without these sites classified as separate sites.

The effect of having Kaposi Sarcoma and Mesothelioma defined as their own sites is that cancers which would otherwise be classified as other sites (lung, for instance) are now not added to those totals. Therefore, the risk of developing or dying of lung cancer for a particular cohort, as determined from a database with Kaposi Sarcoma and Mesothelioma separated out, will be less than the risk calculated for the same cohort in a database without these sites separately listed.

For more information, see SEER Site Recode ICD-O-3 (1/27/2003) DefinitionExternal Web Site Policy andSEER Cause of Death Recode 1969+ (9/17/2004)External Web Site Policy.

[Return to top]

3. Why don't the numbers generated by the current version of DevCan match those generated by previous versions using the same data?

The way DevCan calculates risk estimates changed after version 4.2 as described in Methods Used in DevCan.

[Return to top]

4. How has data provided with DevCan 6.x changed from earlier versions?

There are three main differences in the data provided with DevCan 6.x compared to earlier versions

  1. SEER registry data included with DevCan has changed over the years (see SEER Registry Groupings for AnalysisExternal Web Site Policy for definitions):
    • DevCan 6.6.1 includes the SEER 18 registry database as the default database.
    • DevCan 6.1.0, the SEER 17 registry was the default database.
    • In DevCan 6.0, the SEER 12 registry database was expanded to include the Rural Georgia registry, and is now called SEER 13.
  2. The current version of DevCan uses the November 2011 SEER submission of data. Data included in previous submissions may differ from the current submission due to updates in incidence and population data.
  3. The 0-4 age group used in DevCan 4.2 has been split into two groups, that is, zero year olds are in a group separate from the 1-4 year old group.

[Return to top]

5. What does the follow-back year define?

Please see the explanation of follow-back year.

[Return to top]

6. Why won't the data I used in DevCan versions 4.x work with version 5.1 or later?

Each record DevCan reads from the database file is associated with an age group. In earlier versions of DevCan, this age group was always five years. The format for determining which age group belonged to a particular record was through an index. An index of 1 indicated the record was for ages 0 to 5, an index of 2 for 5 to 10, etc. Beginning with DevCan 5.0, age groups are not restricted solely to five-year intervals. The value used for a particular record is now the actual age at the start of the interval referred to by that record. A record with the age value of 1 indicates the record applies to the interval starting at age 1. The upper end of the interval is determined by the start age of the next record. For example, consecutive records with start ages 0, 10, 25, 50 describe a dataset where the first record is from the interval 0 to 10, the second from 10 to 25, the third from 25 to 50, and the last refers to ages 50+.

A conversion utility was designed to change databases from the format readable by DevCan 4.x to the format used by DevCan versions 5.1 and later. This utility is available for download as a self-extracting zip file: dcdbconv.exe

[Return to top]

7. How do I import data from SEER*Stat into DevCan?

See the SEER*Stat importing tutorials.

[Return to top]

8. Is there a suggested citation for DevCan?

A citation for DevCan, indicating the software version, is recommended. Select About from the DevCan Help menu to view the specific citation. The citation for the software version and database is also provided in a footnote included with DevCan results.