Evidence-based research, epidemiology and alcohol policy: a critique

To demand 'evidence-based medicine' and 'evidence-based research' has become almost a dogma for researchers and policy-makers in Western cultures. A critical analysis shows that the original intention of the term 'evidence-based' was to aim to use the best existing evidence (emphasising that research should be planned and interpreted in a methodologically correct way), while recognising that, in some areas, experimental research would be neither sensible nor feasible. However, as the paradigm has become established, three dubious and commonly coexisting additional connotations have become prominent in practice: 'proven beyond doubt', 'exclusively relying on experimental research' and 'any conclusions based on empirical data'. Drawing on examples from renowned publications that demand an 'evidence-based alcohol policy', this article argues that much of what is presented as 'evidence-based policy' should more appropriately be labelled 'policy-based evidence'.