Jacqueline K. Kueper, Ph.D., Winston Liaw, M.D., M.P.H., Daniel J. Lizotte, Ph.D., and Sian Hsiang-Te Tsuei, M.D., Ph.D., C.C.F.P.
Primary care is the foundation of the health care system and a core driver of population health. Machine learning (ML) has the potential to enhance primary care by improving efficiency, quality, and accessibility, but ML development and implementation in primary care remains limited compared with other medical specialties. There are rich opportunities for collaboration and activity in this area. Effective integration requires methods tailored to the unique characteristics of primary care, including its core functions � first contact, comprehensiveness, coordination, and continuity � as well as the distinct nature of electronic medical record data. Key considerations include using representative primary care data, constructing cohorts that reflect whole-person care, aligning target outcomes with primary care objectives, and developing validation strategies suited to decentralized clinical settings. Addressing these challenges requires interdisciplinary collaboration and increased investment in primary care-specific ML research. Optimizing ML for primary care could enhance clinical decision-making, improve patient outcomes, and drive meaningful innovation in health care.