Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 8369380

Random-effects models, for longitudinal data using Gibbs sampling.

W R Gilks·C C Wang·B Yvonnet·P Coursaget
View original source ↗Export metadata ↓

Abstract

Analysis of longitudinal studies is often complicated through differences amongst individuals in the number and spacing of observations. Laird and Ware (1982, Biometrics 38, 963-974) proposed a linear random-effects model to deal with this problem. We propose a generalisation of this model to accommodate multiple random effects, and show how Gibbs sampling can be used to estimate it. We illustrate the methodology with an analysis of long-term response to hepatitis B vaccination, and demonstrate that the methodology can be easily and effectively extended to deal with censoring in the dependent variable.

Explore related subjects

Follow-Up StudiesHepatitis BHepatitis B Surface AntigensHepatitis B VaccinesHumansImmunizationImmunization, SecondaryInfantLongitudinal StudiesMathematicsModels, StatisticalMultivariate AnalysisJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

W R Gilks, C C Wang, B Yvonnet, P Coursaget. 1993. Random-effects models, for longitudinal data using Gibbs sampling.. https://pubmed.ncbi.nlm.nih.gov/8369380/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1999

Epidemiology and natural history of schizophrenia.

The present review explores the descriptive epidemiology of schizophrenia. Risk factors and correlates are divided into three groups based on whether the available evidence is consistent and strong, consistent and potentially strong, or inconsistent. The paper then considers epidemiologic studies of the course of illness, including a description of findings from the Suffolk County Mental Health Project. Given renewed attention to the need for preventive interventions for individuals at high risk for developing a psychotic illness, epidemiologic values have become more and more central to the conduct of clinical research.

Follow-Up Studies↗
PubMed1999

Three year mortality in heart failure patients with very low left ventricular ejection fractions.

BACKGROUND: Left ventricular ejection fraction is a predictor of the outcome in patients with chronic heart failure. Some treatments cause a small increase in ejection fraction and may, thereby, improve prognosis. METHOD: We studied 99 patients with heart failure and very low left ventricular ejection fractions (all < or =20%). Seventy-four patients had ejection fractions of 11-20% and 25 had < or =10%. Patient follow up was censored at three years. RESULTS: Mortality was 74% at three years. Left ventricular ejection fraction was not a predictor of mortality (P = 0.36). In contrast, peak VO2 at the beginning of the study was a strong predictor of outcome (P = 0.002). CONCLUSION: Three year survival is low when ejection fraction is very low. However, once the ejection fraction is < or =20% ejection fraction is no longer a predictor of mortality. These results suggest that it is unlikely that small increases in ejection fraction will be associated with a survival benefit in this group, and treatments aimed at increasing peak VO2 may be more appropriate.

Follow-Up Studies↗