Under-reporting of Validation Efforts for Health Economic Models Persists Despite the Availability of Validation Tools: A Systematic Review

Objective In this study we aimed to identify possible changes over time in validation efforts and the way in which they are reported for model-based health economic (HE) evaluations, given the introduction of several new validation tools and methods in the past decade. Methods A systematic review was conducted using PubMed and Embase on published HE models for early breast cancer (EBC) for the period 2016 to 2024. AdViSHE-consisting of four validation categories that cover the main HE model aspects-was utilized to systematically evaluate the reported evaluation efforts. The percentage of studies reporting validation per category was compared with the review by de Boer et al. that covers the years 2008 to 2015. Results Of the 2199 records, 78 studies fulfilled the eligibility criteria. Reported validation efforts did not significantly improve compared with the previous period, except for the validation of input data by experts. Reporting on the validation of the conceptual model remained low with around 10% of the studies providing validation. Validation of the computerized model and validation against outcomes using alternative input data were the most underreported validation categories with < 4% of the studies. The validation of model outcomes, specifically cross validity and the comparison with empirical data, remained the most reported categories in this review also, with 52% and 36%, respectively. When validation efforts were reported, this was done in a non-systematic manner, and the tests and results were rarely detailed. Conclusion Overall reporting of validation efforts for model-based HE evaluations in the past decade did not significantly change compared with the previous decade.