Optimising Health Informatics Outcomes--Getting Good Evidence to Where it Matters.

This editorial is part of a For-Discussion-Section of Methods of Information in Medicine about the paper "Evidence-based Health informatics: How do we know what we know?", written by Elske Ammenwerth [1]. Health informatics uses and applications have crept up on health systems over half a century, starting as simple automation of large-scale calculations, but now manifesting in many cases as rule- and algorithm-based creation of composite clinical analyses and 'black box' computation of clinical aspects, as well as enablement of increasingly complex care delivery modes and consumer health access. In this process health informatics has very largely bypassed the rules of precaution, proof of effectiveness, and assessment of safety applicable to all other health sciences and clinical support systems. Evaluation of informatics applications, compilation and recognition of the importance of evidence, and normalisation of Evidence Based Health Informatics, are now long overdue on grounds of efficiency and safety. Ammenwerth has now produced a rigorous analysis of the current position on evidence, and evaluation as its lifeblood, which demands careful study then active promulgation. Decisions based on political aspirations, 'modernisation' hopes, and unsupported commercial claims must cease - poor decisions are wasteful and bad systems can kill. Evidence Based Health Informatics should be promoted, and expected by users, as rigorously as Cochrane promoted Effectiveness and Efficiency, and Sackett promoted Evidence Based Medicine - both of which also were introduced retrospectively to challenge the less robust and partially unsafe traditional 'wisdom' in vogue. Ammenwerth's analysis gives the necessary material to promote that mission.