Painting a portrait: Analysis of national health survey data for cancer genetic counseling.

Background: Despite a growing body of literature describing the geographic and sociodemographic distribution of cancer genetic testing, work focused on these domains in cancer genetic counseling is limited. Research describing the epidemiology of cancer genetic counseling has mainly focused on isolated populations, a single gender (women) and a single condition (hereditary breast and ovarian cancer). Study findings to date are contradictory, making it unclear what, if any, disparities in receipt of cancer genetic counseling exist.

Methods: Utilizing the 2015 National Health Interview Survey (NHIS)-a cross-sectional, in person interview survey collecting self-reported health data for the US population-geographic and sociodemographic factors were compared between those receiving genetic counseling and the national sample. Bivariate analysis and subsequent multivariable logistic regression were performed with stratification by cancer status (affected/unaffected). Reason for (eg, doctor recommended) and focus of (eg, breast/ovarian) genetic counseling were also assessed. To generate nationally representative estimates, all analyses were adjusted for survey weights.

Results: An estimated 4.8 million individuals in the United States had cancer genetic counseling. On bivariate analysis, there were significant differences in proportions undergoing genetic counseling by sex, race/ethnicity, insurance, citizenship, education, age, and cancer status (P < 0.01). After adjustment, however, only female sex (Odds Ratio [OR]: 1.78 [95% CI: 1.18-2.67]) remained a significant predictor of genetic counseling among the affected. Among the unaffected, female sex (OR: 1.70 [1.30-2.21]), non-Hispanic black race (OR: 1.44 [1.02-2.05], reference: non-Hispanic white), graduate education (OR: 1.76 [1.03-2.98], reference: less than high school), and age (OR: 1.06 [1.01-1.11]) predicted higher rates of genetic counseling. An estimated 2.1 million individuals have undergone genetic counseling focused on breast/ovarian cancer, 1.3 million on colorectal cancer, and 1.4 million on "other" cancers. Of those receiving genetic counseling focused on breast/ovarian cancer, 3% were male and 97% female (breast cancer alone-4% male, 96% female); for colorectal cancer, 49% male and 51% female, and for "other" cancers, 60% male and 40% female. The majority of individuals receiving genetic counseling reported they did so because their doctor recommended it (66%), with smaller proportions describing self (12%), family (10%), or media (5%) influences as the primary reason.

Conclusion: This is the first study to depict the sociodemographic and geographic distribution of cancer genetic counseling at the national level. Despite perceived disparities in access, cancer genetic counseling in the United States appears to be accessed by individuals of diverse racial/ethnic backgrounds, with various insurance coverage and educational levels, and across a broad range of ages and geographic regions. The only sociodemographic factor that independently predicted receipt of genetic counseling across both the affected and unaffected population was sex. With physician recommendation as the predominant driver for counseling, targeting physician education, and awareness is crucial to utilization.