Cluster analysis for repeated data with dropout: Sensitivity analysis using a distal event

J Biopharm Stat. 2018;28(5):983-1004. doi: 10.1080/10543406.2018.1428612. Epub 2018 Jan 29.

Abstract

Degeneration of the aortic wall becomes life-threatening when the risk of rupture increases. Cluster analysis on repeated measures of the diameter of the artery revealed two subgroups of patients included in a surveillance program. These results were obtained under the assumption of missingness at random. In this article, we study the vulnerability of the cluster analysis results - the estimated trajectories and the posterior membership probabilities - by applying different missing-data models for non-ignorable dropout, as proposed by Muthen et al. (2011) to the growth of the diameter of the artery.

Keywords: Distal event; incomplete data; latent-class growth models; pattern-mixture models; selection models; sensitivity analysis.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aorta / diagnostic imaging
  • Cluster Analysis
  • Data Interpretation, Statistical*
  • Humans
  • Longitudinal Studies
  • Patient Dropouts / statistics & numerical data*
  • Population Surveillance* / methods