Data mining in oncology: The ukCAT project and the practicalities of working with routine patient data

摘要

Randomised controlled trials (RCTs), and meta-analyses of multiple trials, are considered the pinnacle of medical scientific evidence [1] that should underpin the treatment guidelines used in clinical practice. However, for some time now there has been concern around the generalisability of clinical trial results to the general patient population (e.g. [2–4]). Trial selection criteria, designed to eliminate confounding factors from analysis, favour younger and healthier patients and therefore data are scarce for the elderly and for patients with multiple comorbidities.