Currently, real-world evidence-based comprehensive value evaluation of medicines in medical insurance (hereafter referred to as “R-VALUES”) is gradually becoming an important tool for medicine value assessment and medical insurance decision-making support. However, compared with randomized controlled trials, real-world studies are more susceptible to factors such as selection bias, confounding factors, information bias, and missing data. Therefore, scientific and appropriate research questions and study designs are essential to ensure the reliability of research findings. At present, there are still deficiencies in the study design phase of R-VALUES, which may affect the scientific validity of the generated evidence and its value in decision-making applications. By reviewing research protocols for R-VALUES, this study identifies and summarises the key issues in current evaluation practices. The findings indicate that the central challenge in current research design lies in a disconnect between the design of these studies and the core objectives of R-VALUES. On this basis, this study proposes targeted approaches for optimisation to promote the standardised conduct of R-VALUES and provide higher-quality evidence to support evidence-informed medical insurance decision-making.