OBJECTIVES: Clinical decision support systems (CDSS) are crucial tools in modern healthcare, aiming to improve patient outcomes by providing evidence-based recommendations at the point of care. Traditionally, CDSS relied on pre-programmed guidelines. However, the integration of real-world data (RWD) from various sources is revolutionising CDSS development, enabling more personalised, data-driven decision-making.

METHODS: To develop our RWD CDSS we conducted a targeted literature search, evaluating commercial EHR databases, commercial claims datasets and applicable disease registries. We also explored PubMed and Google Scholer to evaluate existing applications and approaches for CDSS development.

RESULTS: The targeted literature search shows that our proposed approach to CDSS can analyse patients’ EMR data and suggest a treatment plan based on similar cases with successful outcomes. The patient claims data can provide insights into population-level trends. For example, analysis of claims data reveals that a specific chemotherapy regimen leads to better survival rates among breast cancer patients. The CDSS will utilize this data to recommend treatment options. Disease-specific registries can track detailed data on specific patient populations. Consider a cancer registry that tracks immunotherapy outcomes for melanoma patients. The CDSS uses this data to predict the likelihood of response in new patients.

CONCLUSIONS: The future of RWD-powered CDSS is promising, with the integration of artificial intelligence and machine learning for more precise predictions and personalised treatment plans. Ensuring interoperability between healthcare IT systems is essential for wider CDSS adoption. Data quality, standardisation and privacy remain crucial considerations. Robust data security and adherence to privacy regulations are paramount. By harnessing the power of RWD, CDSS can evolve into dynamic tools that empower clinicians to make informed decisions, personalise care and improve healthcare outcomes.