The effect of missing data mechanisms on methods for analyzing the missing data in medical device trials
碩士 === 國立臺北大學 === 統計學系 === 99 === Medical devices are health care products distinguished from drugs for regulatory purposes in most countries based on mechanism of action. Unlike drugs, medical devices operate via physical or mechanical means and are not dependent on metabolism to accomplish their p...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2011
|
Online Access: | http://ndltd.ncl.edu.tw/handle/91988078639365992187 |
Summary: | 碩士 === 國立臺北大學 === 統計學系 === 99 === Medical devices are health care products distinguished from drugs for regulatory purposes in most countries based on mechanism of action. Unlike drugs, medical devices operate via physical or mechanical means and are not dependent on metabolism to accomplish their primary intended effect. The medical device user-is a variable unique to medical device studies and can be responsible for the greatest degree of variability in the clinical outcomes. The approach using last observation carried forward (LOCF) has been used for many years in several therapeutic areas, but has been severely criticized recently. The approach using a likelihood-based multivariate normal linear model is now being used in some areas, but is still considered contentious in others; it has been given the name 'mixed model for repeated measures' (MMRM). Tipping-point analysis as an alternative for handling missing data. Tipping-point analysis provides useful information regarding the treatment effect. This article focuses on comparing these methods.
|
---|