Ethical issues related to comparative effectiveness research, or research that compares existing standards of care, have recently received considerable attention. In this paper we focus on how Ethics Review Committees should evaluate the risks of comparative effectiveness research. We discuss what has been a prominent focus in the debate about comparative effectiveness research, namely that it is justified when “nothing is known” about the comparative effectiveness of the available alternatives. We argue that this focus may be misleading. Rather, we should (...) focus on the fact that some experts believe that the evidence points in favor of one intervention, whereas other experts believe that the evidence favors the alternative. We will then introduce a case that illustrates this point, and based on that, discuss how ERCs should deal with such cases of expert disagreement. We argue that ERCs have a duty to assess the range of expert opinions and based on that assessment arrive at a risk judgment about the study under consideration. We also argue that assessment of expert disagreement is important for the assignment of risk level to a clinical trial: what is the basis for expert opinions, how strong is the evidence appealed to by various experts, and how can clinical trial monitoring affect the possible increased risk of clinical trial participation. (shrink)

The debate about the ethics of the Surfactant, Positive Pressure, and Oxygenation Randomized Trial (SUPPORT) often focuses on the assumptions made by the different parties involved, failing to note the lack of a necessary connection between those assumptions and the main criticism of the study—that the parents appear to have been poorly informed. The fact that the target ranges of oxygen saturation (SpO2) used in SUPPORT were within the range recommended as an appropriate “standard of care” does not mean that (...) the infants randomized to one of two restricted SpO2 ranges received the same treatment they would have received outside of the trial. The corresponding observation that randomization altered the potential harms to which an individual infant was exposed does not mean that the risks of being in SUPPORT were greater than or in addition to the risks associated with usual care. The argument that parents should have been better informed about these potential harms does not entail the (incorrect) assumption that neonatal clinicians make individualized treatment decisions rather than follow a standardized protocol when adjusting the level of oxygen a preterm infant receives. The fact that parents are unaware of the trade‐off clinicians make between these potential harms when establishing such a protocol does not justify the failure to adequately inform parents about the purpose of SUPPORT and the risks and harms that were involved. (shrink)