Primary Care Providers’ Beliefs and Recommendations and Use of Screening Mammography by their Patients

Jennifer S. Haas, William E. Barlow, Marilyn M. Schapira, Charles D. MacLean, Carrie N. Klabunde, Brian L. Sprague, Elisabeth F. Beaber, Jane S. Chen, Asaf Bitton, Tracy Onega, Kimberly Harris, Anna N. A. Tosteson, on behalf of the PROSPR (Population-based Research Optimizing Screening through Personalized Regimens) consortium

Revised breast cancer screening guidelines have fueled debate about the effectiveness and frequency of screening mammography, encouraging discussion between women and their providers. To examine whether primary care providers' (PCPs') beliefs about the effectiveness and frequency of screening mammography are associated with utilization by their patients. Cross-sectional survey data from PCPs (2014) from three primary care networks affiliated with the Population-based Research Optimizing Screening through Personalized Regimens (PROSPR) consortium, linked with data about their patients' mammography use (2011-2014). PCPs (n = 209) and their female patients age 40-89 years without breast cancer (n = 30,233). Outcomes included whether (1) women received a screening mammogram during a 2-year period; and (2) screened women had >1 mammogram during that period, reflecting annual screening. Principal independent variables were PCP beliefs about the effectiveness of mammography and their recommendations for screening frequency. Overall 65.2% of women received >1 screening mammogram. For women 40-48 years, mammography use was modestly lower for those cared for by PCPs who believed that screening was ineffective compared with those who believed it was somewhat or very effective (59.1%, 62.3%, and 64.7%; p = 0.019 after controlling for patient characteristics). Of women with PCPs who reported they did not recommend screening before age 50, 48.1% were nonetheless screened. For women age 49-74 years, the vast majority were cared for by providers who believed that screening was effective. Provider recommendations were not associated with screening frequency. For women ≥75 years, those cared for by providers who were uncertain about effectiveness had higher screening use (50.7%) than those cared for by providers who believed it was somewhat effective (42.8%). Patients of providers who did not recommend screening were less likely to be screened than were those whose providers recommended annual screening, yet 37.1% of patients whose providers recommended against screening still received screening. PCP beliefs about mammography effectiveness and screening recommendations are only modestly associated with use, suggesting other likely influences on patient participation in mammography.