Missing Data in Orthopedic Implant Clinical Trials: A case for Sensible LOCF (Last Observation Carried Forward)
View Presentation *Jianxiong (George) Chu, FDA Keywords: Along with other single imputation methods (simple median/mean, regression mean, minimum or maximum), LOCF is often criticized for its statistical shortcomings to handle missing data in clinical trials (e.g., under-estimated variance and biased estimate). In this talk, I would like to present a case study involving an orthopedic implant, in which the strong assumption for LOCF may be reasonable for certain patients who missed the final primary evaluation at 2 years post-operation, and thus be served as an valuable approach along with other approaches (e.g., multiple imputation assuming missing at random). I will also emphasize that, within a single trial, patients with missing data due to different known reasons could be classified into several categories of most plausible missing mechanisms with clinical input and handled with different approaches accordingly (i.e., not necessary treat all missing data uniformly using either LOCF or multiple imputation).
|
Key Dates
-
April 30 - May 22, 2013
Invited Abstract Submission Open -
June 4, 2013
Online Registration Opens -
August 9 - August 23, 2013
Invited Abstract Editing -
August 23, 2013
Short Course materials due from Instructors -
August 26, 2013
Housing Deadline -
September 9, 2013
Cancellation Deadline and Registration Closes @ 11:59 pm EDT -
September 16 - September 18, 2013
Marriott Wardman Park, Washington, DC