The right to withdraw from research without penalty is well established around the world. However, it has been challenged in some corners of bioethics based on concerns about various harms—to participants, to scientific integrity, and to research bystanders—that may stem from withdrawal. These concerns have become particularly salient in emerging debates about the ethics of controlled human infection (CHI) studies in which participants are intentionally infected with pathogens, often in inpatient settings with extensive follow‐up. In this article, I provide support (...) for preserving the right to withdraw from research without penalty and demonstrate that it is also typically justified in the specific context of CHI studies. The right is well aligned with individual freedoms outside the research setting, where autonomous individuals are permitted to engage in behaviors that will foreseeably cause them harm; where they cannot be compelled to satisfy contracts for their services, nor penalized for failure to do so; and where their behavior is not constrained by public health authorities except in extreme circumstances. These freedoms are supported by U.S. law, as well as by ethical analysis that is more globally relevant. The problems associated with the right to withdraw, however, remain. The best approach to addressing them is not to restrict the right but rather to avoid initiating research when withdrawal would be especially problematic. If research proceeds, steps can still be taken to minimize participant withdrawal without infringing the right. Investigators can avoid participant surprise through informed consent focused on a study’s most burdensome aspects and promote study completion through financial incentives. Should participants nonetheless seek to withdraw, investigators may attempt to persuade them not to do so by encouraging consideration of the range of potential harms that may result. Researchers conducting CHI studies and other research from which withdrawal might be especially problematic should prepare for the possibility of participant withdrawal, respect participant requests to withdraw without penalty, and incorporate various measures to avoid such requests. (shrink)

Research subjects do not always conform to research requirements. When their personal interests conflict with the demands of participation, some subjects surreptitiously break the rules. These subjects are subversive — they undermine the research endeavor. In rejecting the restrictions research imposes, subversive subjects diminish the value of research results. From one vantage point, subversive subjects engage in unethical behavior. They create risks to themselves and others; they also disregard ethical responsibilities to adhere to research agreements and tell the truth. At (...) the same time, subversive subjects expose ethical problems in the design and conduct of clinical trials. Features of the research environment create fertile ground for subject subversion. Intensified policing and guidance are two common strategies for reducing subject subversion, but collaborative reforms are more consistent with the partnership model of clinical research. (shrink)

Scientific reports about clinical research appear objective and straightforward. They describe a study's findings, methods, subject population, number of subjects, and contribution to existing knowledge. The overall picture is pristine: the research team establishes the requirements of study participation and subjects conform to these requirements. Readers are left with the impression that everything was done correctly, by the book.In other places, however, one finds a different and messier picture of clinical research. In this picture, research subjects deviate from the prescribed (...) plan. One author contrasted the “tidy graphics” and “crisp prose” of the New England Journal of Medicine's HIV/AIDS trial publications with reports that subjects shared medications and broke other trial rules. Awareness of this behavior, he wrote, could lead insiders to “conclude that knowledge was resting on something rather less solid than bedrock.”. (shrink)