Application of Affinity Analysis Techniques on Diagnosis and Prescription Data

This study performs an Affinity Analysis ondiagnosis and prescription data in order to discover cooccurrencerelationships among diagnosis and pharmaceuticalactive ingredients prescribed to different patient groups. Theanalysis data collected during consecutive visits of 4,473 patients in a 3 years period, focused on patients suffering byhypertension and/or hypercholesterolemia and appliedassociation rule and sequential rule mining techniques. Thefindings have been validated in the specific dataset usingstatistical analysis methods. Association rule mining shows an association between gastrooesophagealreflux and the medicines prescribed forhypertension and heart diseases, which agrees with findings inthe related literature. Another interesting finding, not yet beenreported in related studies is the association between heartdiseases, gastroesophageal reflux and insulin-dependentdiabetes mellitus for patients that have both hypertension andhypercholesterolemia. Apart from the medical findings, which must be subject offurther research we propose a methodology for the analysis ofdata collected from a continuous screening process of a groupof patients. With the use of data mining techniques we are ableto extract and formulate the potential research questions, which are then validated using statistical methods and can alsobe validated in larger population studies.