Biomedical subjects
Publications and source records attributed to W R Gilks.
We describe Bayesian hierarchical-spatial models for disease mapping with imprecisely observed ecological covariates. We posit smoothing priors for both the disease submodel and the covariate submodel. We apply the models to an analysis of insulin Dependent Diabetes Mellitus incidence in Sardinia, with malaria prevalence as a covariate.
BACKGROUND: In recent years diagnostic practice in psychiatry has become increasingly structured in an attempt to standardize definitions of disorders and improve reliability. At the same time there has been an increasing recognition of the need to take account of uncertainty in the process of diagnostic decision making. For the most part, diagnosis is still represented by a binary outcome while this is known to entail a substantial loss of information. Many diagnostic schemes involve, in part, taking thresholds on the numbers of symptoms required from symptom lists. METHODS: A model is proposed here, using ideas derived from latent class analysis to permit generalization from these schemes through moving from a binary to a probabilistic measure of psychiatric case status and replacing thresholds with smoothed transitions. RESULTS: An outcome measure is produced where disorder status is expressed in terms of probabilities without changing the meaning of the original measure. Prevalence estimates (using ICD-10 Depressive Episode criteria) are more stable and can be given with increased precision. CONCLUSIONS: Disorder status when expressed in this way retains more diagnostic information and provides a useful extension to traditional binary analyses when looking at prevalence and risk factor estimation.
BACKGROUND: Reliable prevalence and risk estimation of psychiatric disorder is a cornerstone to achieving objectives in public health psychiatry. Research strategies have increasingly depended, therefore, upon the progressive evolution and refinement of diagnostic approaches designed to reflect better current knowledge concerning prognosis, course and outcome but essentially the need to improve agreement between users of the various schemes. METHODS: This paper contrasts a conventional with a probabilistic approach to the diagnosis of depression based upon the OPCS United Kingdom National survey of psychiatric morbidity. The probabilistic approach, while designed to mimic current diagnostic practice in relation to the depressive disorders, naturally includes provision for the allocation of respondents on a scale of diagnostic uncertainty according to the severity of their presenting condition. RESULTS: Findings are reported arising from the application of the probabilistic method to three areas of research interest in public health psychiatry, namely; an evaluation of additivity of event exposure and depressive morbidity, secondly use of the approach for investigating psychosocial models of depressive disorder and thirdly for assessing the agreement between depressive disorder when classified according to competing diagnostic schemes. CONCLUSIONS: The results show application of the probabilistic approach to provide a firm basis for achieving gains in both the stability and precision of risk profile estimation for depressive conditions.
Explore the source record for details and available documents.
Bone resorption and formation are coupled both in time and space and may occur simultaneously in the same remodeling unit. A number of studies have shown that the formative phase of the remodeling sequence may undergo temporary interruptions prior to completion and it is possible that bone resorption may be subject to similar interruptions. We have investigated this hypothesis by studying the distribution of eroded depth in resorption cavities in human cancellous bone. Eroded depth was assessed in iliac crest cancellous bone from 41 normal healthy subjects using a cubic spline curve fitting technique. The distribution of mean eroded depths was skewed to the right. Comparison of the observed distribution with an expected distribution, which was calculated from previously published data and assumes resorption begins rapidly and slows as it approaches completion, showed a significantly greater proportion of shallower cavities than expected (p<0001). Similarly, comparison of observed and uniform distributions, which assumes a constant rate of resorption throughout the erosion period, also showed a significantly greater proportion of smaller cavities (p<0.01). In subjects aged less than 39 years, there were fewer small cavities than in those aged 40-59 years. In addition, there was some evidence that females of 40-59 years had a proportionately greater number of smaller cavities than males; however, there were no differences in other age groups. Our results demonstrate a significantly greater proportion of smaller resorption cavities than would be expected from current models of bone remodeling and are consistent with the hypothesis that resorption undergoes temporary interruptions and/or permanent arrest during the process of bone remodeling.
Quantification of the average and interindividual variation in pharmacokinetic behavior within the patient population is an important aspect of drug development. Population pharmacokinetic models typically involve large numbers of parameters related nonlinearly to sparse, observational data, which creates difficulties for conventional methods of analysis. The nonlinear mixed-effects method implemented in the computer program NONMEM is a widely used approach to the estimation of population parameters. However, the method relies on somewhat restrictive modeling assumptions to enable efficient parameter estimation. In this paper we describe a Bayesian approach to population pharmacokinetic analysis which used a technique known as Gibbs sampling to simulate values for each model parameter. We provide details of how to implement the method in the context of population pharmacokinetic analysis, and illustrate this via an application to gentamicin population pharmacokinetics in neonates.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The main aim of the statistical analysis of data collected in the Third International IALSC Workshop on Lung Tumor and Differentiation Antigens, was to identify groups of monoclonal antibodies (MAbs) having similar profiles of reactivity against a variety of cell types in flow cytometry, histology, immunofluorescence and immunocytochemistry experiments. This was achieved through cluster analysis. We describe the methods used in the cluster analysis, and in the data processing leading to it.
Explore the source record for details and available documents.
We construct a unifying representation of the structure of measurement error problems with particular reference to situations commonly encountered in epidemiological studies, and outline how estimation of the parameters of interest can be carried out in a Bayesian framework using Gibbs sampling. We show how this approach can be implemented for designs involving continuous measurement errors assessed through a validation substudy, and discuss our results on simulated data.
Risk factors used in epidemiology are often measured with error which can seriously affect the assessment of the relation between risk factors and disease outcome. In this paper, a Bayesian perspective on measurement error problems in epidemiology is taken and it is shown how the information available in this setting can be structured in terms of conditional independence models. The modeling of common designs used in the presence of measurement error (validation group, repeated measures, ancillary data) is described. The authors indicate how Bayesian estimation can be carried out in these settings using Gibbs sampling, a sampling technique which is being increasingly referred to in statistical and biomedical applications. The method is illustrated by analyzing a design with two measuring instruments and no validation group.
The backcalculation method has been extensively used in AIDS modelling and forecasting. Knowledge of reported AIDS cases, information on the time between HIV infection and onset of AIDS, and assumptions on the rate at which infections occurs, can be used to reconstruct the past history of the HIV epidemic, as well as to provide short term predictions of AIDS incidence. Uncertainty in the three components of the backcalculation method and the increasingly available information on HIV prevalence must be taken into account in order to provide realistic projections. In this paper we discuss ways of acknowledging uncertainty and suggest a Bayesian formulation of the backcalculation idea as a means of combining into a single model both random and systematic variation as well as prior information.
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.
Epidemiological studies of disease can make use of ancillary risk-factors, acquired from individuals outside the disease study. For example, several disease studies might use the same job-exposure matrix to quantify risks due to occupational exposure to industrial agents. We construct a graphical model to combine a logistic regression disease model with models for the ancillary data and the risk-factor distribution in the population. We estimate the graphical model using Gibbs sampling, and in simulations compare it with methods of direct substitution into logistic regression.
Studies have shown that to maintain protection against infection after a primary course of hepatitis B immunisation, revaccination can be scheduled on the basis of an anti-hepatitis B virus surface antigen (anti-HBs) titre obtained 1 month after the booster dose. However, schemes which require post-booster testing may present practical difficulties. We applied a random-effects regression model to data from 118 Senegalese infants given three injections of hepatitis B vaccine about 6 weeks apart and a booster injection at 13 months, and show that revaccination can be scheduled on the basis of an anti-HBs titre recorded at the time of the booster dose. We also show that titre-at-booster is no less accurate in predicting future titre than 1-month post-booster titre. In several other studies the post-booster decline in anti-HBs conforms to the same mathematical description, indicating the generality of our findings.
We show how to estimate expected waiting time to transplantation and match grade potential for each patient awaiting kidney transplantation on a multicenter waiting list. We predict that some easily well-matched patients may be unlikely to receive a well-matched graft through accepting an earlier offer of a poorly matched kidney. Other patients, who are difficult to match, may be unlikely to receive even a poorly matched kidney within a reasonable time.
Methods of data collection for the 2nd Small Cell Lung Cancer Workshop are described, and data reliability is reviewed. The method of cluster analysis of the workshop antibodies is described and discussed. Of the 27,111 results submitted 20,705 were judged to be reliable for analysis and 13,802 of these came from immunohistology experiments. Data derived from immunocytochemistry experiments were somewhat less reproducible than flow cytometry, immunohistology and ELISA experiments. The cluster analysis was developed from methods employed in the leucocyte antigens workshops. Several checks on the methods of cluster analysis and the transformation of data did not substantially alter the final groupings. The workshop confirms that, although there are some methodological difficulties, the cluster analysis can successfully be applied to data derived largely from immunohistology, and thus has applicability to other tumour types.