Abstract
The abundance and availability of health care data has risen exponentially in recent time, coming from a variety of health systems. These data have the ability to provide rich insight into the outcomes of care, standards of care and specifically the descriptive patterns of treatments. However the barrier to this rich insight is twofold; technical and methodological. This methods piece will demonstrate how the application of high performance computing technology has been able to erode technical and methodological boundaries to these data and provide fresh insight in to the chronic disease management within longitudinal patient records. Data from MarketScan, Clinical Practice Research Database (CPRD), IMS-Health and the Hospital Episodes Statistics (HSCIC) have been loaded into a High performance computing centre, hosting >100 Million patient lives with >15 years of linked longitudinal data. Algorithms to define treatment & diagnostic exposures/duration and procedural observations, defined with specific structure query languages. These views feed a visual analytics interface of prevalence/incidence/demographic breakdown/patient interactions and specifically patient pathways. The combination of standardised visual analytics interfaces on these health care data highlight nuances and intricacies in longitudinal patient pathways and records. These automated technologies enable the researchers to instantaneously assess and triangulate trends of disease presentation, treatment and laboratory assessments. We highlight the use of these tools to assess factors used in a Bayesian modelling exercise to assess fluctuations of a laboratory result recorded over time on the prediction of a rare event. Generation and analytics of disease cohorts and from high dimensional datasets requires careful longitudinal interrogation of factors prior to analysis. This factor analysis is often missed in the creation of disease segments and often leads to misinterpretation of data. The use of visual analytics tools combined with high performance computing will drive up the quality and efficiency of pharmacoepidemiological database studies.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
