Using Structured Query Language (SQL) to Stratify a Large-scale De-identified Patient Database: Addressing the top-down versus step-up treatment debate in Inflammatory Bowel Disease (IBD)
Date of Award
Master of Professional Studies
Big Data; Comparative Effectiveness Research; Inflammatory Bowel Disease; Step-up versus Top-down; Structured Query Language
This Capstone Project addresses the debate between 'step-up' and 'top-down' treatment approaches common in Inflammatory Bowel Disease (IBD). In light of the ineffectiveness of conventional clinical research, the Project establishes another method of research using modern techniques in data science to build patient identification algorithms and separate the patient population into cohorts based on similar prognostic factors. Researchers can study the comparative effectiveness of alternative treatment approaches by querying de-identified disease-specific databases using a string of structured query language (SQL) commands. Patient cohorts are constructed according to initial treatment pathway, and a comparison is drawn between prescribing frequency of 'step-up' medications (5-ASAs and corticosteroids) and frequency of 'top-down' medications (immunomodulators and biologics) from a de-identified database of 26,000 patients.
Sinderbrand, Matthew T., "Using Structured Query Language (SQL) to Stratify a Large-scale De-identified Patient Database: Addressing the top-down versus step-up treatment debate in Inflammatory Bowel Disease (IBD)" (2014). University College: Healthcare Leadership Capstones. 28.