Programmatic assessment and competency development in postgraduate medical education: a systematic review and narrative synthesis

Abstract

Introduction Programmatic assessment (PA) is widely endorsed as a way to organize assessment around competency development in postgraduate medical education (PGME), yet no systematic review has synthesized the empirical evidence in this setting. We therefore synthesized the evidence on its contribution to competency development in PGME and on the factors that shape its implementation.Methods A systematic review was conducted following PRISMA 2020 guidelines (PROSPERO: CRD420261349850). We searched MEDLINE, Embase, Scopus, Web of Science, and EBSCOhost from January 2005 onward. Each included study was mapped against four operational features of PA derived from van der Vleuten et al. Methodological quality was assessed using the Mixed Methods Appraisal Tool (MMAT), and findings were synthesized using a narrative approach.Results Twenty studies met inclusion criteria. Four reported competency-related outcomes, all with observational designs. Two evaluated purposefully designed PA systems, and two analyzed milestone-based data generated within regulated competency-based training without an explicitly designed PA system. These studies reported increases in competency ratings and improved examination outcomes over training, though none could establish a causal link to the assessment system. The remaining 16 studies focused on implementation processes, stakeholder perceptions, and measurement properties. Challenges including feedback documentation barriers, misalignment between intended and perceived assessment stakes, and variable stakeholder engagement were reported consistently across settings. Methodological quality was uneven across the included studies (nine rated High, six Moderate, and five Low on a descriptive MMAT-based quality index).Discussion The empirical evidence for PA in PGME remains limited relative to its theoretical foundations. Most evidence reflects how stakeholders experience the system or whether assessment tools produce reliable data, rather than whether the system improves competency or patient care. Future research should extend outcome measurement beyond system-internal metrics and account for the contextual specificity of PA across diverse educational and regulatory environments.