Statistics in medicine
-
Statistics in medicine · Dec 2002
Modelling tooth emergence data based on multivariate interval-censored data.
Studies on emergence of (permanent) teeth are published regularly in the dental literature. Besides descriptive statistics (mean or median values) on emergence times, comparisons between boys and girls are of interest. Gender comparisons are intersubject analyses, but also intrasubject questions, like 'Is there a left-right symmetry with respect to the mean (median) emergence times?' are of interest. ⋯ We will extend a GEE-type test proposed by Huster et al. for bivariate right-censored data to the multivariate setting with interval-censored data. Central to our paper is to provide appropriate statistical models to resolve some dental questions on emergence. The analyses are based on data from the longitudinal Signal-Tandmobiel study.
-
Statistics in medicine · Oct 2002
Comparative StudyPerformance of weighted estimating equations for longitudinal binary data with drop-outs missing at random.
The generalized estimating equations (GEE) approach is commonly used to model incomplete longitudinal binary data. When drop-outs are missing at random through dependence on observed responses (MAR), GEE may give biased parameter estimates in the model for the marginal means. A weighted estimating equations approach gives consistent estimation under MAR when the drop-out mechanism is correctly specified. ⋯ Weighted GEE resulted in smaller finite sample bias than GEE. However, when the drop-out model was misspecified, weighted GEE sometimes performed worse than GEE. Weighted GEE with observation-level weights gave more efficient estimates than a weighted GEE procedure with cluster-level weights.
-
Meta-analyses involving the synthesis of evidence from cluster randomization trials are being increasingly reported. These analyses raise challenging methodologic issues beyond those raised by meta-analyses which include only individually randomized trials. In this paper we review and comment on a selected number of these issues, including problems of study heterogeneity, difficulties in estimating design effects from individual trials and the choice of statistical methods.
-
In 1990 the International Conference on Harmonization (ICH) effort was begun with the intent of standardizing the drug registration and approval process. The need to rationalize and harmonize regulation was driven by concerns over rising costs of health care, escalation of the cost of research and development and the need to meet the public expectation that there should be a minimum delay in making safe and efficacious new treatments available to patients in need. Since most regulatory agencies have limited resources to interact with sponsor companies, standardized guidelines would help expedite communications with companies at the programme design stage. ⋯ If the guidelines do not gain a solid foothold early on, then drift between the regions in use of the guidelines will defeat the goals of the ICH. While the ICH covers the entire drug development process, this paper will review the guidelines that pertain most closely to clinical trials and their use in the drug registration process. Some of the guidelines have been approved and some are still in the development stage.
-
Statistics in medicine · Sep 2002
Intensive short courses in biostatistics for fellows and physicians.
At both of our universities we teach (with colleagues) introductory courses in statistics for fellows and physicians. We do not expect that those taking these courses will be able to do their own statistical work, but rather the intention is for them to 'learn the language' and to facilitate future collaboration. ⋯ We will discuss the factors (what works and what does not) that may contribute to a successful course, a comparison to other courses, and our self-evaluation strategy. Finally, we will cover the financial arrangements that we have made when teaching these courses.