Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Mixed models for longitudinal left-censored repeated measures.

Rodolphe Thiébaut1, Hélène Jacqmin-Gadda

  • 1ISPED, INSERM E0338 Biostatistics, Université Victor Segalen Bordeaux II, 146, Rue Léo Saignat 33076, Bordeaux Cedex, France. rodolphe.thiebaut@isped.u-bordeaux2.fr

Computer Methods and Programs in Biomedicine
|May 12, 2004
PubMed
Summary

Handling left-censored data in longitudinal studies, like Human Immunodeficiency Virus (HIV) viral load, requires advanced methods. This study reviews likelihood-based approaches for accurate statistical analysis, avoiding biased results from simple imputation.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Immunogenicity and protective efficacy on non-adjuvanted CD40-targeting SARS-CoV-2 vaccines in non-human primates.

EBioMedicine·2026
Same author

Heterogeneous trajectories of executive function in clinical Alzheimer's disease: Results from the 3C study.

Alzheimer's & dementia (Amsterdam, Netherlands)·2026
Same author

Cross-reactive Bundibugyo antibody responses after licensed Ebola vaccines.

medRxiv : the preprint server for health sciences·2026
Same author

Gene set analysis for time-to-event outcome: comparison of a new approach based on the generalized Berk-Jones statistic with existing methods in presence of intra gene-set correlation.

Briefings in bioinformatics·2026
Same author

CD40.HIVEnv, an antibody mediated vaccine, induces long-term and recall immunogenicity in non-HIV-1 infected volunteers.

Scientific reports·2026
Same author

A regularized multi-state model for covariate selection with interval-censored survival data.

Biometrics·2026

Area of Science:

  • Biostatistics
  • Epidemiology
  • Medical Statistics

Background:

  • Longitudinal studies frequently encounter left-censored repeated measures, particularly in biomedical research.
  • Accurate quantification of outcomes like Human Immunodeficiency Virus (HIV) plasma viral load is often limited by assay detection limits.
  • Standard imputation methods for left-censored data can introduce bias in statistical estimations and their standard errors.

Purpose of the Study:

  • To review and present likelihood-based methods for handling left-censored outcomes in linear mixed models.
  • To demonstrate the application of these methods using SAS Proc NLMIXED.
  • To compare the utility and limitations of different software programs for analyzing such data.

Main Methods:

  • Review of statistical literature on handling left-censored data in longitudinal models.

Related Experiment Videos

  • Application of linear mixed models using SAS Proc NLMIXED.
  • Comparative analysis of statistical software capabilities for left-censored data.

    • Main Results:

    • Likelihood-based methods provide a statistically sound approach to managing left-censored repeated measures.
    • SAS Proc NLMIXED can effectively fit linear mixed models with left-censored outcomes.
    • Different software packages exhibit varying capabilities and limitations in handling left-censored data.

    Conclusions:

    • Appropriate statistical methodologies are crucial for unbiased analysis of longitudinal data with left-censoring.
    • Likelihood-based approaches and tools like SAS Proc NLMIXED offer robust solutions for HIV viral load and similar research.
    • Understanding software limitations is essential for reliable scientific conclusions in studies with detection limits.