individualized informed consent usually associated with research. Conversely, some activities that are routinely carried out by epidemiologists do raise ethical issues that may benefit from careful scrutiny or even reconsideration, even if they have long traditions and are sanctioned by regulations or statutes. CIOMS International Ethical Guidelines for Epidemiological Studies, 2009 Introduction Guidance point 2 Patient safety activities, even when they do not meet the definition of research requiring ethical review, may involve more than minimal risk to patients and health-care providers in some situations. It is the responsibility of those working in patient safety activities to be aware of ethical issues and seek guidance as needed. Distinguishing between research and programme implementation is not always straightforward. For example, even though patient safety programmes are typically designed to improve care within a particular organization, their results may be disseminated to other organizations. Just as with research, the benefits of patient safety activities may ultimately extend beyond the population that was exposed to the risks. Moreover, the methodologies of many patient safety initiatives closely resemble those used in research, especially in cluster designs, including alterations to standard practice, use of randomization, and the systematic collection of data. Thus, as CIOMS has recognized in the context of epidemiology, and the Ottawa Statement in the context of cluster studies,12 even activities that do not formally meet the definition of research may benefit from undergoing “careful ethical scrutiny or even reconsideration.” Based on this principle, patient safety activities that do not constitute research may benefit from ethical oversight or advice whenever those activities involve greater-than-minimal risks. For activities that do not formally meet the definition of research, this risk assessment may be made by a reference group that is independent of the activity and who are well versed in both ethics and the principles of patient safety improvement. Some institutions may choose to give this review authority to an REC. For those who may wish to publish the results of non-research activity, it would be advisable to obtain advice from a REC prior to starting the activity. Regardless of an activity’s label, the most important question to consider when undertaking patient safety activities is whether the patients and providers who participate in these programmes will be subjected to any risks they would otherwise not assume. Pronovost P, Needham D, Berenholz S, et. al. An Intervention to Decrease Catheter Related Bloodstream Infections in the ICU. N Engl J Med, 2006, 355:272. This patient safety project was implemented to determine the effectiveness of an evidence-based intervention aimed at reducing rates of catheteracquired bloodstream infections in ICUs. The intervention aimed at improving physicians’ use of five procedures that have been previously shown to reduce infections. These procedures include proper hand washing, full barrier precautions during insertion of the catheter, cleaning the patient’s skin with chlorhexidine, avoiding the femoral site if possible, and removing unnecessary catheters. Agreement to participate was obtained from 108 ICUs from 67 hospitals in the state of Michigan, United States. Each of the ICUs implemented the evidence-based intervention, and infection rates before the implementation of the intervention were compared with infection rates after its implementation. Each ICU was responsible for reporting aggregate infection data back to the project staff. The project staff developed the intervention and conducted the evaluation but were not members of any of the ICUs involved in the study. The study was submitted to the REC of the home institution of the project staff before implementation. Indeed, confusion by experienced professionals about whether projects like this should be classified as research vs. quality improvement practice led to public controversy around this particular case, and questioning from government authorities, highlighting the need for clearer guidance around when projects such as this should be subject to ethical oversight and