Background Throughout the world, the number and quality of head CT and MRI scans is increasing. While this proliferation of imaging will ideally result in earlier and more accurate diagnoses, it will inevitably result in higher numbers of incidental findings. Unruptured intracranial aneurysms, longstanding developmental anomalies such as arachnoid cysts or Chiari malformations, or non-malignant brain tumours are just some of the entities that may be detected incidentally. In the US and the UK, for example, rates of non-malignant meningioma are increasing. In both settings, we found that the increase was driven by increases in tumours classified as ‘radiographically diagnosed’, which tend to be smaller tumours not requiring immediate intervention. These findings provide evidence of the potential impacts of changing imaging practices.

We have identified four points along the diagnostic continuum whereby policy and/or interventions could be made to mitigate the risks over overdetection, overdiagnosis, and overtreatment: (1) the threshold for scanning and number of scans; (2) the fidelity of the imaging modality (e.g. the thickness of CT scan slices and the differing sensitivity of MRI and CT to discriminate particular pathologies) (3) diagnosis and risk communication; (4) management of incidental findings.

Methods In the first third of the workshop (30 minutes), we will hear from experts about the state of evidence of benefits and harms of brain imaging. In the second third, participants will form groups according to the four themes enumerated above and prioritize key research questions addressing overdiagnosis within those themes. Finally, participants will brainstorm methods to answer one of their key research questions.

Objectives Using a framework of the diagnostic pathway, the objectives of this workshop are to: Discuss and identify the most pressing research questions related to brain imaging and overdetection/overdiagnosis; Identify methods to pursue these questions; create a virtual interest group to spark & sustain new multidisciplinary collaborations. Following the workshop, volunteers from the group will convene to produce a commentary for publication summarizing the key conclusions.

Chair & Panelists Diana Withrow (Chair).

Senior Research Fellow, Nuffield Department of Primary Care Health Sciences, University of Oxford, UK.

Relevant research interests Cancer epidemiology, early detection and diagnosis.

Amy Berrington de Gonzalez (Panelist).

Professor & Team Lead, Clinical Cancer Epidemiology, The Institute of Cancer Research, London, UK.

Relevant research interests Late effects of medical radiation.

John Brandt Brodersen (Panelist).

Professor & General Practitioner, Department of Public Health, University of Copenhagen, Denmark.

Relevant research interests Psychosocial consequences of false-positive test results and overdiagnosis.

Andrew Solomon (Panelist).

Associate Professor & Neurologist, University of Vermont Medical Center, Burlington, Vermont, USA.

Relevant research interests Definition and overdiagnosis of multiple sclerosis.