The ethical labour of overdiagnosis research: Negotiating care in early detection.

Particularly in relatively well-funded healthcare systems, medicine increasingly seeks to detect disease before it causes symptoms. Early detection can prevent suffering and premature death, but it can also diagnose conditions that will never become harmful. Although the existence of overdiagnosis is supported by clinical and epidemiological literature, its implications for practice remain contested. This paper examines how clinicians and researchers decide what evidence of overdiagnosis should mean for good care and how they act on it (or not). Drawing on three years of ethnographic fieldwork at conferences and seminars on overdiagnosis and early detection in the UK and internationally, supplemented by formal interviews with attendees and many informal conversations, I examine how evidence about the benefits and harms of intervention is interpreted, debated and translated into practice. Using Barrett's oesophagus as a central case, I seek to make a conceptual contribution to the literature by identifying two overlapping ethical orientations found in these fields. An "ethics of doing" treats intervention as a responsible response to the possibility of future disease. An "ethics of restraint" treats limiting intervention as a way of protecting people from harms produced by diagnosis and treatment. These are not fixed positions held by opposing groups: the same people may move between them as they respond to different forms of risk, responsibility and uncertainty. I argue that evidence alone does not determine how evidence of overdiagnosis should be acted upon. The same uncertainty can support intervention, because clinicians cannot know in advance who will develop serious disease, or restraint, because they cannot know who will benefit from intervention. What is at stake is therefore not simply whether overdiagnosis exists, but how harms and responsibilities are understood. By treating overdiagnosis as an ethnographic problem, the paper shows how competing responses to early detection are made consequential in practice and produce divergent notions of care.