bMultivariate logistic regression was adjusted for age educa
bMultivariate logistic regression was adjusted for age, education, household income, employment condition, chronic disease, and self-rated health.
FHCA, family history of cancer; KoGES, Korean Genome and Epidemiology Study.
Table 4. Comparison of Health Behaviors According to FHCA, Including FHCA by Cancer Type in Females Without Cancer (KoGES 2004−2013) 400
Note: Boldface indicates statistical significance (p<0.05) compared to the reference groups. aReference group was those with no cancer diagnosis by a physician and also with no FHCA at the time of survey. bMultivariate logistic regression was adjusted for age, education, household income, employment condition, chronic disease, and self-rated health. FHCA, family history of cancer; KoGES, Korean Genome and Epidemiology Study.
lifestyle modifications.17 This is partly caused by the fact that SYBR Safe DNA Gel Stain current guidelines for management of high-risk individuals primarily include medical options, while somewhat neglecting lifestyle modifications.34,35 Studies have shown that people with an FHCA of the breast or colorectum were more likely to comply with healthcare providers’ recommendations to improve health behav-iors, but the absolute number of people who received those recommendations was low.36,37 In the current study population, females with an FHCA showed a higher education level than those without an FHCA. Smoking rate is known to be higher in more educated women than in those with less education.38 Men are
known to consume more cigarettes than women, but women have greater difficulty in quitting smoking.39,40
In addition, a previous study reported that women who recently experienced FHCA did improve their behaviors related to physical activity, diet, and alcohol consump-tion, but did not quit smoking.41 Females’ perception of cancer risk because of FHCA was greater than that for males, but the perceived causes of cancer regarding smoking and alcohol drinking were much lower than that for males.42 Also, Koreans are known to have strong bonds among family members, and a close familial rela-tionship is regarded as more important in older females. However, a sufficiently large sample size might have resulted in statistically significant differences, despite limited clinical significance in the current study, such as the small, albeit statistically significant, difference in female smoking rates between those with no FHCA and those with any FHCA (2.3% vs 2.6%).
When stratified by cancer type of FHCA, males with an FHCA of the thyroid showed different health behav-ior patterns, which was worse than those with other types of FHCA. Thyroid cancer has been the most com-mon cancer in Korea since 2008, and a large portion of Korean thyroid cancers is suspected to be from overdiag-nosis,43 suggesting possibly different patient characteris-tics compared with other types of cancer. In addition, males with an FHCA of the thyroid showed different demographic and socioeconomic characteristics com-pared with those with other types of FHCA. Although these characteristics were adjusted for in multivariate analysis, overdiagnosed thyroid cancers in family mem-bers may still result in different health behavior patterns for those with an FHCA of the thyroid, compared with those with other types of FHCA.
Some limitations of this study should be considered. First, as the participants of HEXA were health examinees resid-ing in metropolitan areas or major cities, generalization to the entire Korean population may be limited. In addition,
because this study only includes participants aged more than 40 years, the authors were unable to identify FHCA’s effect on health behaviors in the younger population. Sec-ond, information regarding FHCA was collected from a self-reported questionnaire. The structure of Korean fami-lies has become more nuclear, and thus, exact FHCA sta-tus may be inaccurate. However, previous studies have shown that an individual’s reported FHCA, especially in the case of first-degree relatives, is highly reliable.44,45 Additionally, the Korean government offers the National Cancer Screening Program, which provides free cancer screening or a 10% co-payment. Hence, many Koreans are well informed not only of their own cancer status but also of other family members, and minimal information bias is expected. Third, this study is unable to demonstrate any cause and effect relationships between FHCA and health behaviors, as HEXA is cross-sectional data and health behaviors before and after FHCA were not avail-able. Fourth, when health behaviors were compared by cancer type, nine groups were compared simultaneously. When a set of statistical inferences are tested simulta-neously, multiple-testing correction may be needed, which was not considered here. However, not adjusting for mul-