Abstract Details
Activity Number:
|
503
|
Type:
|
Invited
|
Date/Time:
|
Wednesday, August 7, 2013 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract - #307013 |
Title:
|
The Regression Discontinuity Design in Epidemiology: An Application to Statins
|
Author(s):
|
Sara Geneletti and Gianluca Baio*+
|
Companies:
|
London School of Economics and Political Science and University College London
|
Keywords:
|
causality ;
local average treatment effect ;
statins
|
Abstract:
|
It is important to determine if a drug is being prescribed at the point in disease progression where it will be most beneficial to patients. Determining prescription guidelines falls to expert clinicians. The regression discontinuity (RD) design provides a principled way of evaluating whether a particular prescription rule is successful. Many drugs are prescribed according to guidelines, e.g. statins, a class of cholesterol-lowering drugs. In the UK, the NHS recommends that patients are prescribed statins when their risk of developing CVD in the subsequent 10 years, as calculated by a risk calculator, exceeds 20%. We exploit this guideline to estimate the effect of the treatment around the threshold. We apply the RD to statin prescription in THIN, a UK database which contains a representative sample of anonymised patient records. A number of Bayesian models are explored, ranging from simple regressions to cubic penalized splines. We compare results to those obtained from clinical trials of statins and consider the underlying assumptions in the context of the data. Of particular interest is the issue of "weak instruments" as doctors do not always follow prescription guidelines.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.