Guidance point 1 Any patient safety activity that constitutes research, regardless of its methodology, should be submitted to a Research Ethics Committee (REC). For purposes of REC oversight, patient safety activities constitute research when: ñ they are aimed at addressing a specific question; and ñ they use a predefined approach or method for collecting data in response to the question they intend to address; and ñ their findings are intended to be applied to settings beyond those in which the activity or programme is implemented. International guidelines define research as a systematic activity designed to develop or contribute to generalizable knowledge. Before conducting any study, researchers should consider whether the research is warranted and whether the selection and recruitment of participants is fair. There should be good reason to explore safety practice in research beyond routine evaluation. Furthermore, vulnerable populations should not be differentially exposed to any extra risks brought up by the research without good reason to avoid exacerbating pre-existing inequalities.10 While health research has the potential to bring benefits to society and to the research communities, it can be associated with risks to the research participants. It is the obligation of those responsible for conducting the research to protect participants by minimizing those risks and seeking approval of an independent Research Ethics Committee (REC). The rationale for requiring third party review of research is that the participants are often asked to undertake risks that are not offset by any potential personal benefit, in order to develop scientific knowledge that may benefit others in the future. REC review is designed to ensure that the risks assumed by the participants are carefully 2. 11Interpreting existing guidance examined and minimized to the extent that is reasonably possible, that the risks that remain are reasonable in relation to the potential benefits of the study, and that other ethical considerations, such as informed consent and confidentiality, are adequately addressed. Patient safety research may entail potential risks to both patients and health-care providers, and like any other research activity must be submitted to a research ethics committee for review. Patient safety research must be distinguished from activities designed to provide benefits to an existing patient population, including activities characterized as “programme implementation or practice,” such as surveillance or quality improvement. Such patient safety activities are in many instances mandated by regulatory or administrative authorities or units of health-care organizations, and are typically undertaken to serve the interests of the individuals who are cared for in these same organizations. In general, current regulations do not require these activities to be reviewed by an REC. For instance, a hospital may implement a patient safety programme in which data are systematically collected from medical charts of surgery patients to better understand hospital-specific circumstances leading to infection (for example, staffing patterns, scheduling of procedures, or availability of equipment). The hospital’s goal is to identify specific aspects of its organization or delivery of care aiming to reduce the risk of infection. When conducted as part of quality improvement practice, such activities are generally expected to quickly change clinical protocols and practices at the institution in question. Although patient data will be collected in a systematic way, the programme’s focus on hospitalspecific circumstances means that the data collected are unlikely to be generalizable beyond that particular setting. As such, many institutions would characterize the activity as quality improvement and thus, not require external ethics review, despite the fact that identifiable human data are being collected. On the other hand, if a project abstracts data from medical charts in a series of hospitals to determine how infection rates can be reduced among hospitals more broadly, so that hospitals can learn more widely from the information, the project would generally fall within accepted definitions of research requiring REC review. In practice, however, there is a great