The relative efficiency of time-to-threshold and rate of change in longitudinal data

M. C. Donohue, A. C. Gamst, R. G. Thomas, R. Xu, Laurel A Beckett, R. C. Petersen, M. W. Weiner, P. Aisen

Research output: Contribution to journalArticle

19 Scopus citations


Randomized, placebo-controlled trials often use time-to-event as the primary endpoint, even when a continuous measure of disease severity is available. We compare the power to detect a treatment effect using either rate of change, as estimated by linear models of longitudinal continuous data, or time-to-event estimated by Cox proportional hazards models. We propose an analytic inflation factor for comparing the two types of analyses assuming that the time-to-event can be expressed as a time-to-threshold of the continuous measure. We conduct simulations based on a publicly available Alzheimer's disease data set in which the time-to-event is algorithmically defined based on a battery of assessments. A Cox proportional hazards model of the time-to-event endpoint is compared to a linear model of a single assessment from the battery. The simulations also explore the impact of baseline covariates in either analysis.

Original languageEnglish (US)
Pages (from-to)685-693
Number of pages9
JournalContemporary Clinical Trials
Issue number5
StatePublished - Sep 2011



  • Linear mixed models
  • Longitudinal data
  • Marginal linear models
  • Power
  • Survival analysis

ASJC Scopus subject areas

  • Pharmacology (medical)
  • Medicine(all)

Cite this

Donohue, M. C., Gamst, A. C., Thomas, R. G., Xu, R., Beckett, L. A., Petersen, R. C., Weiner, M. W., & Aisen, P. (2011). The relative efficiency of time-to-threshold and rate of change in longitudinal data. Contemporary Clinical Trials, 32(5), 685-693.