BackgroundPrecision medicine has been adopted in a range of clinical settings where omics data have led to greater characterisation of disease and stratification of patients into subcategories of phenotypes and pathologies. However, in orthopaedics, precision medicine lags behind other disciplines such as cancer. Joint registries have now amassed a huge body of data pertaining to implant performance which can be broken down into performance statistics for different material types in different cohorts of patients. The National Joint Registry of England, Wales and Northern Ireland (NJR) is now one of the largest datasets available. Other registries such as those from Sweden and Australia however contain longer follow-up. Together, these registries can provide a wealth of informative for the orthopaedics community when considering which implant to give to any particular patient.Questions/PurposesWe aim to explore the benefits of combining multiple large data streams including joint registries, published data on osteoarthritis (OA) pathogenesis and pathology and data concerning performance of each implant material combination in terms of biocompatibility. We believe that this analysis will provide a comprehensive overview of implant performance hopefully aiding surgeons in making more informed choices about which implant should be used in which patient.MethodsData from three joint registries were combined with established literature to highlight the heterogeneity of OA disease and the different clinical outcomes following arthroplasty with a range of material types.ResultsThis review confirms that joint registries are unable to consider differences in arthritis presentation or underlying drivers of pathology. OA is now recognised to present with varying pathology with differing morbidity in different patient populations. Equally, just as OA is a heterogeneous disease, there are disparate responses to wear debris from different material combinations used in joint replacement surgery. This has been highlighted by recent high-profile scrutiny of early failure of metal-on-metal total hip replacement (THR) implants.ConclusionsBringing together data from joint registries, biomarker analysis, phenotyping of OA patients and knowledge of how different patients respond to implant debris will lead to a truly personalised approach to treating OA patients, ensuring that the correct implant is given to the correct patient at the correct time.Electronic supplementary materialThe online version of this article (doi:10.1007/s11420-016-9531-7) contains supplementary material, which is available to authorized users.