FBNSMCAN is a two-digit count variable.
00: Not in Universe
03: Beginning in 2015, top code of 3 or more
21: Prior to 2015, top code of 21 or more
98: Unknown-Not Ascertained
99: Unknown-Don't Know
For sample adults with at least one full brother who ever had stomach cancer (FBSMCAN), FBNSMCAN reports the number of full brothers who had ever been told they had stomach cancer. Only persons with at least one full brother (FBNUM) who ever had cancer of any kind (FBNHCAN) were asked follow-up questions about what kind(s) of cancer their full brother(s) had, and about the number of full brothers they had with the specified type(s) of cancer (such as stomach cancer).
FBNSMCAN is one of a series of related variables in the "Family History" section of the 2000 and later Cancer Modules. In all years, this section begins with the general comment:
According to the 2000 and 2005 Field Representative's Manuals, the purpose of this and other family cancer history questions is to "determine the prevalence of cancer in families in the population" to "allow a greater understanding of how data obtained from genetics clinics can be applied to the general population." The 2010 and later Manuals provide a similar rationale, stating that the collection of these data "will allow analysts a greater understanding of how family history relates to cancer."
The Manuals for all years define "full siblings" as people who have "the same biological father and mother." Thus, a respondent's full brother is a brother who has the same biological father and mother as the respondent (i.e., not half-brothers, stepbrothers, adopted brothers, foster brothers, or brothers-in-law).
Although FBNSMCAN is substantively comparable across all years, users should note two changes in the recoding of values that may affect analyses of data from multiple cancer supplement years.
The first, and most recent, change affects the top code value for FBNSMCAN. Prior to 2015, FBNSMCAN was top coded at 21. Starting in 2015, FBNSMCAN was top coded at 3. The second change, affecting how missing data values were treated for sample adults who completed only part of the cancer supplement, took place between the 2000 and 2005 cancer supplement. In 2000, persons who answered less than 50 percent of the questions in the supplement had a value of "not ascertained" automatically inserted in the appropriate fields of the Cancer Control Module file, whereas in 2005 and later years, persons who answered none of the questions in the supplement had a value of "not ascertained" automatically inserted in the appropriate fields of the Cancer Control Module file.
- 2000; 2005; 2010; 2015: Sample adults aged 18 and older whose biological brother(s) ever diagnosed with stomach cancer.
- 2000, 2005, 2010, 2015
- 2000, 2005, 2010, 2015 : SAMPWEIGHT