Assessment of Validation of Health-Economics Decision Models In Intervention Studies of Seasonal Influenza and Breast Cancer

Objectives: We aimed to review recently published health-economic (HE) decision models to assess the reporting of validation efforts. An infectious disease (seasonal influenza, SI) and a chronic disease (breast cancer, BC) were used as examples, giving a preliminary insight in the reporting of validation efforts in the overall HE literature. Methods: A literature search was performed in Pubmed and Embase to retrieve full-text HE modeling studies, published between 2008 and 2014. Type of evaluation, model and intervention were extracted, as well as information on model outcomes, journal and funding. Reporting on model validation was evaluated by checking for the presence of the word validation and its conjugates, and by using AdViSHE, a tool which contains a structured list of relevant items for validation. Results: The literature search resulted in 53 SI and 45 BC studies. In 41 studies (42%) the word validation or its conjugates was mentioned, but only in a small percentage in the context of model validation. The terminology used around validation was found to be ambiguous. Model validation efforts were reported in a minority of studies. However, some studies do show good reporting examples. Cross validation of study outcomes was reported most often, but the quantity and quality of this reporting varied. More validation efforts were reported in BC than in SI. Conclusions: Only a limited number of studies reported on model validation efforts, although it may be assumed that more efforts have been taken than were reported. In particular, the differences between SI and BC may not mean that less efforts were undertaken to validate SI models. Although validation is deemed important by many researchers, this is not reflected in the reporting habits of HE modeling studies. Better reporting of validation efforts would be desirable to further enhance decision-makers' confidence in HE models and their outcomes.