Determining the harm-benefit balance of an intervention: for each patient.

The decision on whether or not to offer a patient a medical, diagnostic, therapeutic or other type of intervention is mainly based on the harm-benefit balance of this intervention for that particular patient. The benefits that matter most are those that correspond to a tangible improvement for the patient rather than improvement in a surrogate end-point. The harms include the various potential or common adverse effects and drawbacks. The harm-benefit balance of an intervention is first evaluated at the population level. Evaluation of the benefits therefore takes into account the strength of the evidence obtained in clinical trials, the magnitude and probability of the benefits in these trials, and the profile of the patients enrolled. Evaluation of the harms involves identifying the drawbacks and amassing a body of evidence to determine potential adverse effects. Evaluation of the adverse effects also takes into account particular situations (age, pregnancy, concomitant diseases and treatments, etc.) and the probability and consequences of error. The harm-benefit balance cannot be reduced to an artificial, fixed mathematical ratio. Its assessment occasionally involves a degree of subjectivity. It is sometimes biased due to manipulation of the data. At the individual level, the harm-benefit balance depends on: the characteristics, objectives and values of each patient; the healthcare professionals involved and the medical and social environment. It is best evaluated in collaboration with the persons concerned, so that it can provide a basis for shared decision-making. The harm-benefit balance of an intervention can change. Its periodic re-assessment, taking into account new evidence and any changes in the patient's situation, provides an opportunity to re-examine the decisions taken, in the patient's best interests.