Introduction

Because individual patients weigh benefits and harms differently, it is important to provide quantitative information about the benefits and harms of screening interventions, particularly when benefits and harms are in equipoise.1 The goal of this article is to help clinicians have a more fulsome understanding of the implications of different screening situations. Guideline developers may also find this helpful when estimating the magnitude of overdiagnosis in screeningwhere the outcome is not overt disease.

Many clinicians discuss the potential benefits and harms of screening with their patients. But without an estimate of their magnitude, it is difficult to engage patients in shared decision making (SDM). One recognised harm of screening is overdiagnosis which ‘means making people patients unnecessarily, by identifying problems that were never going to cause harm’.2 Although its magnitude has been much debated, there is little debate about the existence of overdiagnosis in screening.3 Overdiagnosis estimates are mostly available in screening for overt disease (eg, cancer). Here, we propose a method to make such estimates in the context of screening for disease precursors and risk factors for disease.

Screening can identify non-disease conditions such as the risk of an outcome (eg, risk of a fragility fracture) or potential precursors of disease (eg, colonic polyps). In these instances, the measurement of the extent of overdiagnosis is a challenge that has not yet been clearly addressed. For overdiagnosis in risk assessment, we are not aware of any attempt to define the concept nor provide an estimate of its magnitude. For precursors of disease, some authors have described overdiagnosis in colon cancer screening4 or used modelling to estimate overdiagnosis of precancerous cervical lesions.5 Building on these reflections, we first outline the measurement of overdiagnosis in overt disease, and then propose a conceptual model and a method to …