Small Changes, Big Differences? Cohort Variation by Parameter Specifications in Claims-Based Drug Use Evaluations

Real-world drug utilization and outcome assessments using health insurance claims or other routinely-collected electronic health data have become a common type of study in pharmacoeconomics and pharmacoepidemiology. However, these observational studies may be sensitive to parameter specifications such as outpatient pharmacy dispensing stockpiling algorithm and lead to inconsistent results. This poster examines the impact of small specificaiton changes on comparative risk assessments among drug users in a test case.