1 Cardiothoracic Surgery Division, McGill University Health Centre, Royal Victoria Hospital, Montreal, Quebec, Canada H3A 1A1 2 Oxford International Programme in Evidence-Based Health Care, University of Oxford, Oxford, UK 3 Department of Surgery, McGill University, Montreal, Canada 4 Nuffield Department of Surgical Science, University of Oxford, UK 5 Health Services Research Unit, University of Aberdeen, Aberdeen, UK 6 Centre for Statistics in Medicine, University of Oxford, UK Correspondence to: P L Ergina firstname.lastname@example.org Accepted: 15 March 2013 Cite this as: BMJ 2013;346:f3011 doi: 10.1136/bmj.f3011 The IDEAL framework describes the stages of evaluation for surgical innovations. This paper considers the role of observational studies in the exploration and assessment stages. At the exploration stage, the surgical intervention is usually more widely used, and observational studies should collect prospective data from multiple surgeons, deal with factors such as case mix and learning, and prepare for a de! nitive evaluation at the next stage of assessment. Although a randomised controlled trial is preferable, a high quality observational study would be acceptable if a randomised trial is not feasible or, on rare occasions, deemed unnecessary.