Risk assessments are important tools in the planning of offshore well operations. They ensure that involved crews are prepared for, and aware of, challenges that may occur for a specific operation. The risk assessments may be qualitative as well as probabilistic to serve different needs. In both cases, it is vital to identify and evaluate risk factors that influence the crew's ability for detection and reaction to events that may affect the status of well barriers. The human-machine interface (HMI) is one of the main means of communication between the state of well barriers and the crew, and it is therefore important to incorporate the role of HMI in human reliability assessment (HRA). This article evaluates HMI as a specific risk factor in well risk assessment and HRA. A novel method that combines well accident report reviews and analysis with probabilistic well risk assessment is presented. The method investigates the degree that accident causality descriptions reveal the HMI as a contributing factor, and the results from reviews are then applied in a probabilistic risk model directed to assess the risk of well control loss during an operation. The method is demonstrated with case studies that emulate well risk assessment that incorporate an existing offshore HRA framework. A potential modification and improvement to HRA causal model and checklist procedure is suggested as result of case studies to help ensure that key aspects of the HMI as risk factor not systematically become overlooked. (C) 2017 Elsevier Ltd. All rights reserved.