Probabilities and health risks: A qualitative approach

Health risks, defined in terms of the probability that an individual will suffer a particular type of adverse health event within a given time period, can be understood as referencing either natural entities or complex patterns of belief which incorporate the observer's values and knowledge, the position adopted in the present paper. The subjectivity inherent in judgements about adversity and time Frames can be easily recognised, but social scientists have tended to accept uncritically the objectivity of probability. Most commonly in health risk analysis, the term probability refers to rates established by induction, and so requires the definition of a numerator and denominator. Depending upon their specification, many probabilities may be reasonably postulated for the same event, and individuals may change their risks by deciding to seek or avoid information. These apparent absurdities can be understood if probability is conceptualised as the projection of expectation onto the external world. Probabilities based on induction from observed frequencies provide glimpses of the future at the price of acceptance of the simplifying heuristic that statistics derived from aggregate groups can be validly attributed to individuals within them. The paper illustrates four implications of this conceptualisation of probability with qualitative data from a variety of sources, particularly a study of genetic counselling for pregnant women in a U.K. hospital. Firstly, the official selection of a specific probability heuristic reflects organisational constraints and values as well as predictive optimisation. Secondly, professionals and service users must work to maintain the facticity of an established heuristic in the face of alternatives. Thirdly, individuals, both lay and professional, manage probabilistic information in ways which support their strategic objectives. Fourthly, predictively sub-optimum schema, for example the idea of AIDS as a gay plague, may be selected because they match prevailing Social value systems. (C) 1998 Elsevier Science Ltd. All rights reserved.