This article analyses research exploring medication adherence, help-seeking behaviour, screening and behaviour change to argue that all interventions have the potential for both benefit and harm. Accordingly, health psychology may have inadvertently contributed to psychological harms (e.g. lead times, anxiety, risk compensation and rebound effects); medical harms (e.g. medication side effects, unnecessary procedures) and social harms (e.g. financial costs, increased consultations rates). Such harms may result from medicalisation or pharmaceuticalisation. Or, they may reflect the ways in which we manage probabilities and an optimistic bias that emphasises benefit over cost.