Propensity Score Matching And Time Varying Confounding

1283 Words6 Pages
STUDY 3: PROPENSITY SCORE MATCHING AND TIME-VARYING CONFOUNDING IN A STUDY OF STATINS AND THE RISK OF VENOUS THROMBOEMBOLISM Aims • To evaluate the association between statin use and the risk of VTE controlling for time-varying confounding. • To examine whether this association differs by body mass index (BMI) • To examine various characteristics of exposure including type of statin, dose of statin, and recency/duration of statin use in relation to VTE risk Study design This study will be a population-based cohort study. Study population The study population will consist of all adult patients aged 40-79 years during the period 1995-2015 in the CPRD. I will exclude all patients with a history of cancer (non-melanoma), VTE, cerebrovascular disease, coronary heart disease, heart failure, coagulopathies, vasculitis, and chronic kidney disease prior to cohort entry. Participants will be required to have at least one year of recorded history in the database before the study entry date. The date of entry into the cohort (start date) will be defined as the last of the date patients became 40 years old or 1995, and registered in the CPRD). Because dyslipidemia covers a broad range of lipid abnormalities and statins are widely prescribed to a broad spectrum of people (from relatively healthy, middle age to the older, very sick population with numerous comorbidities) controlling for confounding by indication in this study will be a big challenge. To mitigate this, an option will be

More about Propensity Score Matching And Time Varying Confounding

Open Document