Main
Vast amounts of data are being generated across various domains of human activity, including healthcare and medical insurance. Harnessing this data effectively could bring immense benefits, enabling the identification of risks and positive effects associated with drugs, medical devices, procedures, and more. Big data analysis has the potential to revolutionize decision-making across critical areas such as healthcare, education, and economics.
My website focuses on statistical methods tailored for analyzing large observational databases to uncover associations between drugs and adverse medical conditions. Most of these methods are implemented in SAS, allowing for flexibility in adapting them to new statistical features, diverse databases, and varying data formats. These methodologies can also be extended beyond healthcare to other subject areas, offering broad applicability and utility.
Direct links to statistical methods for mining large observational databases:
Disproportionality Methods for Large Observational Data
Univariate Self Controlled Case Series
Multivariate Self Controlled Case Series
Bayesian Logistic Regression for Large Observational Data