BACKGROUND: Portfolios in post-graduate healthcare education are used to support reflective practice, deliver summative assessment, aid knowledge management processes and are seen as a key connection between learning at organisational and individual levels. This systematic review draws together the evidence on the effectiveness of portfolios across postgraduate healthcare and examines the implications of portfolios migrating from paper to an electronic medium across all professional settings. METHODS: A literature search was conducted for articles describing the use of a portfolio for learning in a work or professional study environment. It was designed for high sensitivity and conducted across a wide range of published and unpublished sources relevant to professional education. No limits for study design or outcomes, country of origin or language were set. Blinded, paired quality rating was carried out, and detailed appraisal of and data extraction from included articles was managed using an online tool developed specifically for the review. Findings were discussed in-depth by the team, to identify and group pertinent themes when answering the research questions. RESULTS: Fifty six articles from 10 countries involving seven healthcare professions met our inclusion criteria and minimum quality threshold; mostly uncontrolled observational studies. Portfolios encouraged reflection in some groups, and facilitated engagement with learning. There was limited evidence of the influence of a number of factors on portfolio use, including ongoing support from mentors or peers, implementation method, user attitude and level of initial training. Confounding variables underlying these issues, however have not been fully investigated. A number of authors explored the reliability and validity of portfolios for summative assessment but reports of accuracy across the disparate evidence base varied. Links to competency and Quality Assurance frameworks have been demonstrated. There were conflicting reports about whether the different purposes of portfolios can be combined without compromising the meaningfulness of the contents. There was good evidence that the flexibility of the electronic format brought additional benefits to users, assessors and organisations, and encouraged more enthusiastic use. Security of data remained a high priority issue at all levels, and there was emerging evidence of successful transfer between electronic portfolio systems. CONCLUSION: The evidence base is extensive, but contains few high quality studies with generalisable messages about the effectiveness of portfolios. There is, however, good evidence that if well implemented, portfolios are effective and practical in a number of ways including increasing personal responsibility for learning and supporting professional development. Electronic versions are better at encouraging reflection and users voluntarily spend longer on them. Regular feedback from a mentor enhances this success, despite competing demands on users' time and occasional scepticism about the purpose of a portfolio. Reports of inter-rater reliability for summative assessments of portfolio data are varied and there is benefit to be gained from triangulating with other assessment methods. There was insufficient evidence to draw conclusions on how portfolios work in interdisciplinary settings.