External validation of a prediction model for timely implementation of innovations in radiotherapy

Background and purpose: The aim of this study was to externally validate a model that predicts timely innovation implementation, which can support radiotherapy professionals to be more successful in inno-vation implementation.Materials and methods: A multivariate prediction model was built based on the TRIPOD (Transparent Reporting of a multivariate prediction model for Individual Prognosis Or Diagnosis) criteria for a type 4 study (1). The previously built internally validated model had an AUC of 0.82, and was now validated using a completely new multicentre dataset. Innovation projects that took place between 2017-2019 were included in this study. Semi-structured interviews were performed to retrieve the prognostic vari-ables of the previously built model. Projects were categorized according to the size of the project; the suc-cess of the project and the presence of pre-defined success factors were analysed.Results: Of the 80 included innovation projects (32.5% technological, 35% organisational and 32.5% treat-ment innovations), 55% were successfully implemented within the planned timeframe. Comparing the outcome predictions with the observed outcomes of all innovations resulted in an AUC of the external validation of the prediction model of 0.72 (0.60-0.84, 95% CI). Factors related to successful implementa-tion included in the model are sufficient and competent employees, desirability and feasibility, clear goals and processes and the complexity of a project.Conclusion: For the first time, a prediction model focusing on the timely implementation of innovations has been successfully built and externally validated. This model can now be widely used to enable more successful innovation in radiotherapy. CO 2023 The Author(s). Published by Elsevier B.V. Radiotherapy and Oncology xxx (2023) xxx-xxx This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).