Random-effects models for multivariate repeated measures

Fieuws, S.; Verbeke, Geert; Molenberghs, G.
October 2007
Statistical Methods in Medical Research;Oct2007, Vol. 16 Issue 5, p387
Academic Journal
Mixed models are widely used for the analysis of one repeatedly measured outcome. If more than one outcome is present, a mixed model can be used for each one. These separate models can be tied together into a multivariate mixed model by specifying a joint distribution for their random effects. This strategy has been used for joining multivariate longitudinal profiles or other types of multivariate repeated data. However, computational problems are likely to occur when the number of outcomes increases. A pairwise modeling approach, in which all possible bivariate mixed models are fitted and where inference follows from pseudo-likelihood arguments, has been proposed to circumvent the dimensional limitations inmultivariate mixed models. An analysis on 22-variate longitudinal measurements of hearing thresholds illustrates the performance of the pairwise approach in the context of multivariate linear mixed models. For generalized linear mixed models, a data set containing repeated measurements of seven aspects of psycho-cognitive functioning will be analyzed.


Related Articles

  • Potential Pitfalls of Clinical Prediction Rules.  // Journal of Manual & Manipulative Therapy (Journal of Manual & Ma;2008, Vol. 16 Issue 2, p69 

    The author reflects on the potential pitfalls of the clinical prediction rule (CPR) which is a useful tool for a selected and discreet population of patients. According to the author, the unrecognized difficulty of CPR is characterized by the relevant failure in meeting the appropriate...

  • Cognition of and Demand for Education and Teaching in Medical Statistics in China: A Systematic Review and Meta-Analysis. Wu, Yazhou; Zhou, Liang; Li, Gaoming; Yi, Dali; Wu, Xiaojiao; Liu, Xiaoyu; Zhang, Yanqi; Liu, Ling; Yi, Dong // PLoS ONE;Jun2015, Vol. 10 Issue 6, p1 

    Background: Although a substantial number of studies focus on the teaching and application of medical statistics in China, few studies comprehensively evaluate the recognition of and demand for medical statistics. In addition, the results of these various studies differ and are insufficiently...

  • Longitudinal analysis methods for multivariate data. Haiqun Lin; Holford, Theodore R. // Statistical Methods in Medical Research;Oct2007, Vol. 16 Issue 5, p383 

    The article discusses various reports published within the issue, including one on high-dimensional multivariate longitudinal responses and another on a latent trait model in the context of multidimensional longitudinal data.

  • Linking HES maternity records with ONS birth records. Abrahams, Carole; Davy, Kevin // Health Statistics Quarterly;Spring2002, Vol. 13, p22 

    This paper describes a pilot study into the feasibility of linking maternity records from the hospital episode statistics (HES) system with the corresponding birth records held by the Office for National Statistics (ONS). If successful this would add to the value of statistics derived from both...

  • Population ideals and sample realities -- why we still need access to comprehensive information about populations. Barrett, Geraldine; Peacock, Janet L. // Journal of Epidemiology & Community Health;Apr2005, Vol. 59 Issue 4, p257 

    The article presents a discussion on population ideals and sample realities in public health research. Public health population research will be rendered impossible if individual consent for all secondary uses of health data becomes the norm. Permissions for medical record linkage are moving...

  • Ignoring Dependency between Linking Variables and Its Impact on the Outcome of Probabilistic Record Linkage Studies. TROMP, MIRANDA; MÉRAY, NORA; RAVELLI, ANITA C. J.; REITSMA, JOHANNES B.; BONSEL, GOUKE J. // Journal of the American Medical Informatics Association;Sep/Oct2008, Vol. 15 Issue 5, p654 

    Objectives: This study sought to examine the differences between ignoring (naïve) and incorporating dependency (nonnaïve) among linkage variables on the outcome of a probabilistic record linkage study. Design and Measurements: We used the outcomes of a previously developed probabilistic...

  • Applying a Medical Budget System to Biostatistics Clinics. Makie, Toshio // Journal of Medical Systems;Oct2004, Vol. 28 Issue 5, p469 

    Biostatistics are indispensable in clinical medicine because doctors increasingly need to publish and understand medical articles that use statistical methods. An opportunity is needed for medical doctors to interact with biostatisticians on statistical issues. This paper's objective is to...

  • Data quality maintenance of the Patient Master Index (PMI): a 'snap-shot' of public healthcare facility PMI data quality and linkage activities. Williams, Kelly; Robinson, Kerin; Toth, Alexandra // Health Information Management Journal;2006, Vol. 35 Issue 1, p10 

    Patient (or person) master index (PMI) data quality activities in public, acute healthcare facilities in the state of Victoria, Australia were evaluated in terms of health information management-information technology best practice including data standards and practice guidelines. The findings...

  • Privacy preserving interactive record linkage (PPIRL). Hye-Chung Kum; Krishnamurthy, Ashok; Machanavajjhala, Ashwin; Reiter, Michael K.; Ahalt, Stanley // Journal of the American Medical Informatics Association;Mar2014, Vol. 21 Issue 2, p212 

    Objective Record linkage to integrate uncoordinated databases is critical in biomedical research using Big Data. Balancing privacy protection against the need for high quality record linkage requires a human--machine hybrid system to safely manage uncertainty in the ever changing streams of...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics