institutions, compared with the patient safety activity not being implemented. 17Interpreting existing guidance self-perceived competency and skills, providers might have potentially exposed some failures of their own or of their team or institution, thereby creating some level of social, psychological or even economic risks.The REC would have to determine, once safeguards have been put in place, whether a study like Case Study 3 poses more than minimal risks to the individuals involved. Some types of studies are usually recognized as posing minimal risk. For example, research where individually identifiable human data are not collected, or when the data of the study is already in the public domain, tends to be considered of minimal risk and in many contexts may be exempt from a REC review altogether.15 Similarly, projects that are randomized, whether at the clinic/institution level, or at the level of individual patients, may be considered minimal risk research if, based on best evidence, no study arms are considered therapeutically experimental and no study arms pose a significant deviation from the care usually received by patients in the study setting. Certain projects may pose minimal risk when implemented in some settings, but may pose greater than minimal risk in other contexts, for example, in settings that are retributive towards providers or where confidentiality protections cannot be guaranteed. It is, therefore, important for those involved in patient safety research to understand the institutional culture, norms, laws and regulations where the research takes place, and identify those laws and regulations which are relevant to their work. For studies that document rates of errors, it may be especially important to know whether any local laws require the reporting of provider error to any authorities or licensing boards, as well as whether any exceptions to such policies exist for patient safety research. The existence of such laws may compromise the neutrality of researchers and pose risks to the participants in the study. Investigators documenting errors should also talk to trusted informants in local settings to know the chance of retribution for providers, if errors are found, as well as providers’ anticipated anxiety about possible retribution. Patient safety professionals must also be sensitive to power relationships that exist between senior and junior hospital staff. In Case Study 3, those involved in the patient safety activity made clear to providers that all data would be kept confidential and therefore that no individual level data would be reported to hospital management. In settings where the local context could increase risks related to reporting or retribution, researchers may need to consider whether it would be ethically preferable to abandon the study, in order to prevent additional risks to participants, or whether it would be effective and feasible to build additional safeguards into the study. In some circumstances, collecting data anonymously (rather than confidentially) may be sufficient to protect against risk of harm. 18Ethical issues in Patient Safety Research Cullen DJ, Sweitzer BJ, Bates DW, et. al. Preventable adverse drug events in hospitalized patients: a comparative study of intensive care and general care units. Crit Care Med, 1997, 25:1289-1297. A patient safety project was implemented at 11 different ICUs and general care units in two tertiary care hospitals in the United States, with the objective of understanding the causes of adverse drug events (ADEs) resulting from human errors in drug use, so that they might be prevented in the future. Data on all ADEs that occurred in any of those 11 units over a six-month period were included in the project. Data on ADEs were collected in several ways: all nurses and pharmacists in those 11 units were asked to report all ADEs to nurse investigators involved in the project, a nurse investigator visited each unit twice a day on weekdays to obtain information from hospital staff on any ADEs that had recently occurred, and each day, the nurse investigator also reviewed all medical charts to collect information on reported and potential ADEs. All data were then reviewed by two physicians. These physicians judged the severity of