Communication for better decisions: Not about ‘risk'

Abstract

Decisions—choices between options/strategies/policies—will better to the extent they incorporate better structural modeling of the scenarios which follow from the adoption of each option; better assessments of the chances of the events and outcomes which are contained in these scenarios; better assessment of the relevant preferences for the possible outcomes (including, very importantly in the health context, intertemporal preferences); and better ways of integrating these probability judgments and value judgments into an overall evaluation of each option. Each of these elements is conceptually distinct and improvement in them requires very different knowledge and skills, as witnessed by the formidable literatures about what ‘better’ means in each. We cannot avoid engaging with them separately by a side-step into a world and language of ‘risk', which is either a synonym for chance, a synonym for harm, or refers to an implicit synthesis of probability judgments and value judgments that inevitably prejudices their separate and explicit assessment. Decision analysis meets the requirement of addressing these elements separately. If, in making health-related choices, people are informal decision analysts whose models include life expectancy and quality of life, then in order to communicate with them we need to be thinking in terms of the multiple and separate tasks of better ‘decision structure communication’ and ‘probability communication’ and ‘life expectancy communication’ and ‘quality of life communication’ and ‘time discount communication’ and ‘option evaluation communication'. Not better ‘risk communication'.