eduzhai > Physical Sciences > Physics Sciences >

Evaluating Methods for Dealing with Missing Outcomes in Discrete-Time Event History Analysis: A Simulation Study

  • Save

... pages left unread,continue reading

Document pages: 41 pages

Abstract: Background: In discrete-time event history analysis, subjects are measured once eachtime period until they experience the event, prematurely drop out, or when thestudy concludes. This implies measuring event status of a subject in each timeperiod determines whether (s)he should be measured in subsequent time periods.For that reason, intermittent missing event status causes a problem because,unlike other repeated measurement designs, it does not make sense to simplyignore the corresponding missing event status from the analysis (as long as thedropout is ignorable). Method: We used Monte Carlo simulation toevaluate and compare various alternatives, including event occurrence recall,event (non-)occurrence, case deletion, period deletion, and single and multipleimputation methods, to deal with missing event status. Moreover, we showed themethods’ performance in the analysis of an empirical example on relapse to druguse. Result: The strategies assuming event (non-)occurrence and therecall strategy had the worst performance because of a substantial parameterbias and a sharp decrease in coverage rate. Deletion methods suffered fromeither loss of power or undercoverage issues resultingfrom a biased standard error. Single imputation recovered the bias issue butshowed an undercoverage estimate. Multiple imputations performed reasonably with a negligiblestandard error bias leading to a gradual decrease in power. Conclusion: On the basis of the simulation results and real example, we provide practicalguidance to researches in terms of the best ways to deal with missing eventhistory data.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×