Evaluating the balance of risks and benefits of medical interventions is fundamental to interpreting both the clinical usefulness and the regulatory approvability of medical interventions. Clinical trials often have insufficient sample size and follow-up time to accurately evaluate harms. This has two implications: 1) the initial evaluation of harms is based on broadly evaluating for signals as well as integrating information from pre-clinical and early clinical testing, and 2) the assessment of harms is not based solely on evaluation of p-values and statistical significance. Therefore it is important to continually asses risks and benefits over the entire life cycle of medical interventions. New information should trigger overall reevaluations of risk and benefits, and the conditions of use which justify risks. The justifiable degree of risk may differ between serious and life-threatening diseases compared to self-resolving disease, and benefits demonstrated in superiority compared to non-inferiority trials.