Primary care physicians' willingness to offer a new genetic test to tailor smoking treatment, according to test characteristics.

Emerging pharmacogenetics research may improve clinical outcomes for common complex conditions typically treated in primary care settings. Physicians' willingness to offer genetically-tailored treatments to their patients will be critical to realizing this potential. According to recent research, it is likely that genotypes used to tailor smoking will have pleiotropic associations with other addictions and diseases, and may have different frequencies across populations. These additional features may pose an additional barrier to adoption. To assess physicians' willingness to offer a new test to individually tailor smoking treatment according to specific test characteristics, we conducted a national mailed survey of 2,000 U.S. primary care physicians (response rate: 62.3%). Physicians responded to a baseline scenario describing a new test to tailor smoking treatment, and three additional scenarios describing specific test characteristics based on published research; there was random assignment to one of two survey conditions in which the test was described as a genetic or non-genetic test. Our findings indicate physicians' self-reported likelihood (0-100 scale) that they would offer a new test to tailor smoking cessation treatment ranged from 69%-78% across all scenarios. Relative to baseline scenario responses, physicians were significantly less likely to offer the test when informed that the same genotypes assessed for treatment tailoring: (1) may also identify individuals predisposed to become addicted to nicotine (p<.001), (2) differ in frequency by race (p<.004), and (3) may have associations with other conditions (e.g., alcohol and cocaine addiction, attention deficit hyperactivity disorder ADHD and Tourette Syndrome) (p<.01). Describing a new test to individually tailor smoking treatment as a "genetic" versus non-genetic test significantly reduced physicians' likelihood of offering the test across all scenarios, regardless of specific test characteristics (p<.0007). Effective education of primary care physicians will be critical to successful integration of promising new pharmacogenetic treatment strategies for smoking.