Factors Associated with Breast MRI Use: A Population-Based Analysis

Abstract

Although controversial, the use of breast magnetic resonance imaging (MRI) is widespread. We sought to determine factors that influenced its use in a population-based sample. The National Health Interview Survey is conducted annually by the Centers for Disease Control and is designed to be representative of the American population. Data from 2010 were queried for the use of breast MRI and associated sociodemographic and risk characteristics. Of the 11,222 women aged ≥30years who were surveyed, 4.7% reported ever having a breast MRI. Nearly a quarter were done as part of a "routine exam" and <5% were done for "family history" or for "high risk." Factors correlating with MRI use on univariate analysis included age, race, personal and/or family history of breast cancer, history of benign breast biopsy, perceived risk, and insurance. On multivariate analysis, African-American race (p=0.001), personal history (p<0.001), history of benign biopsy (p<0.001), and high perceived risk (p<0.001) were significantly associated with increased MRI use. In a cohort without a personal history of breast cancer, race, history of benign biopsy, and perceived risk were independent correlates of breast MRI, whereas family history, age, and insurance were no longer significant. Personal history is the strongest factor associated with breast MRI use. However, whereas race, history of benign biopsy, and perceived risk were independently associated with MRI use, family history was not. These findings call into question whether current practice patterns follow evidence-based guidelines.