|
KMID : 0379919990240010001
|
|
Journal of The Korea Socity of Health Informatics and Statistics
1999 Volume.24 No. 1 p.1 ~ p.6
|
|
|
A Study of the difference between crude failure and net failure probabilities by cause of death
|
|
|
|
|
Abstract
|
|
|
|
In modem society in which chronic diseases are the major causes of death, elimination of the effect of disease are appropriate under the practical situation of other causes acting simultaneously without imposing of elimination of other causes(net probability) rather than under the unrealistic assumption of elimination of other causes(net probability). The net probability is larger than crude probability, given competing risks. From the analysis of simulated data, it was found that the difference between the net probability and crude probability becomes larger as the proportion censored(due to loss-to-follow-up and terrnination of study) gets smaller. If the proportion censored is held constant, the difference becomes greater as the proportion of deaths from the cause of interest(compared to other caeses) gets smaller. An illustrative example was given for prostate cancer patients in Hawaii, and an application of cause-specific failure probabilities was discussed.
|
|
|
KEYWORD
|
|
|
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|