How To Use Propensity Score Analysis（如何使用倾向得分分析）.pdfVIP

How To Use Propensity Score Analysis Lisa Kaltenbach, MS Department of Biostatistics lisa. altenbach@ April 11, 2008 Outline  Bac ground/Motivation  Propensity Score Estimation  Propensity Score Matching  Regression Adjustment/Stratification  Example Code/Analysis  Conclusions  References Motivational Ancedote  Two heart surgeons walk into a room. − The first surgeon says, “Man, I just finished my th 100 heart surgery!”. − The second surgeon replies, “Oh yeah, I finished th my 100 heart surgery last week. I bet Im a better surgeon than you. How many of your patients died within 3 months of surgery? Only 10 of my patients died.” − First surgeon smugly responds, “Only 5 of mine died, so I must be the better surgeon.” − Second surgeon says, “My patients were probably older and had a higher risk than your patients.” Comparing apples to oranges?  There may be important differences in patient characteristics between treatment groups.  Want to show difference in outcome is attributable to difference in treatment (or patient condition) and not due to comparing apples to oranges.  Nonrandomized comparisons give rise to apples-and-oranges scepticism.  Sometimes it is infeasible or unethical to assign patients to different treatments. Purpose of Propensity Scores  Can produce apples-to-apples comparison under some nonrandomized conditions.  Provides a way to summarize covariate information about treatment selection int