Diagnostic investigations: harmless tests or booby traps?

Most people view diagnostic tests as harmless means of obtaining additional information and resolving uncertainties, thereby enabling the treatment of medical problems. Although this is true in most cases, there are situations in which it is not. The aim of the article is to demonstrate how diagnostic tests entail potential harms and benefits, arising from the knowledge they yield and its consequences, which has psychological, moral, and existential dimensions. These harms justify consideration of a strategy we call intentional unknowing, an choice to decline the acquisition of diagnostic knowledge. We show when and how this strategy is appropriate and propose a conceptual model that categorizes tests by the extent to which they provide high or low knowledge benefit and high or low knowledge harm.

The need for shared decision-making is lowest when the potential benefits of diagnostic knowledge are low, and the potential harms are high. When the benefits and harms of diagnostic knowledge are both high, the need for shared decision-making is highest; the default position is collaborative deliberation. When the benefit of diagnostic knowledge is high and the harm is low, the need for shared decision-making depends on the test's accuracy. The model also illustrates when patients might prefer "not to know".

As diagnostic and prognostic capabilities expand, uncertainty also increases, with significant implications for the utility of diagnostic knowledge and, in some situations, for ensuring that patients are informed about the balance of harms and benefits and that their preferences are respected.