The evidence for positive correlations between years of education and health status is consistent. The patterns of causalities underlying these correlations are likely to be complex, but here the focus is upon one direction of causality; namely how education makes us healthy. If education has the potential to improve health outcomes, then policy-makers and practitioners should understand the interrelated processes involved. As a first step towards this understanding, I suggest a typology of intermediate factors linking education to health outcomes, falling under four headings; economic factors, access to health services, health-related practices, and coping with stress. Although the development of the typology was the purpose of the research, the process of doing so led to the identification of a number of policy-relevant issues or themes that apply to all groups of intermediate factors: (1) immediate psychosocial outcomes of learning play a fundamental role in generating the behaviours, skills and personal attributes that have lasting effects upon health; (2) health benefits of education that apply to individuals do not always apply to communities; (3) health benefits of education depend upon the nature of the learning experienced; and (4) they also depend upon the social and economic structures in which the learner is situated.