Noisy nocebo harms: a two-part problem for active drug surveillance

Abstract Objectives To illustrate both the bioethical and data-quality or ‘epistemic’ issues that are constitutive of the ‘nocebo effect’ in active pharmaceutical surveillance. This is intended to both motivate careful risk communication with patients and to assist researchers in developing the growing practice of active pharmacovigilance. Methods An examination and consideration of the definition of nocebo effect and an investigation into an imagined case of how nocebo effects arise in the process of active surveillance with the purposes of illustrating the two-part problem. Key Findings Nocebo effects as negative expectation or conditioning effects are bioethical issues in the course of active pharmacovigilance insofar as they are genuine harms patients can experience. However, this is only one ‘side’ of the problem that nocebo effects entail. When nocebo effects are reported as suspected adverse drug reactions to databases aimed at discovering the actual pharmacological properties of drugs, this is ‘noise’ in those databases. Conclusion Due to the two-part problem that nocebo effects entail in active surveillance, i.e. that they are genuine patient harms and they are ‘noise’ in pharmacovigilance databases aimed at discovering the pharmacological properties of drugs, we ought to be careful to mitigate these in practice and in research methods. This more robust account of the issues that nocebo effects are constitutive of more comprehensively motivates careful risk communication.