Problems in assessing rates of infection with methicillin resistant Staphylococcus aureus.
Explore the source record for details and available documents.
Cross Infection↗
Biomedical subjects
David J Spiegelhalter
Publications and source records attributed to David J Spiegelhalter.
Funnel plots for comparing institutional performance.
'Funnel plots' are recommended as a graphical aid for institutional comparisons, in which an estimate of an underlying quantity is plotted against an interpretable measure of its precision. 'Control limits' form a funnel around the target outcome, in a close analogy to standard Shewhart control charts. Examples are given for comparing proportions and changes in rates, assessing association between outcome and volume of cases, and dealing with over-dispersion due to unmeasured risk factors. We conclude that funnel plots are flexible, attractively simple, and avoid spurious ranking of institutions into 'league tables'.
Adolescent↗
Prior distributions for the intracluster correlation coefficient, based on multiple previous estimates, and their application in cluster randomized trials.
Numerous estimates for the intracluster correlation coefficient (ICC) are available in research databases and publications. When planning a cluster randomized trial, an anticipated value for the ICC is required; currently, researchers base their choice informally on the magnitude of previous ICC estimates. In this paper, we make use of the wealth of ICC information by formally constructing informative prior distributions, while acknowledging the varying relevance and precision of the estimates available. Typically, for a planned trial in a given clinical setting, multiple relevant ICC estimates are available from each of several completed studies. Our preferred model allows for the imprecision in each ICC estimate around its underlying true value and, separately, allows for the similarity of ICC values from the same study. The relevance of each previous estimate to the planned clinical setting is considered, and estimates corresponding to less relevant outcomes or population types are given less influence. We find that such downweighting can increase the precision of the anticipated ICC. In trial design, the prior distribution constructed allows uncertainty about the ICC to be acknowledged, and we describe how to choose a design that provides adequate power across the range of likely ICC values. Prior information on the ICC can also be incorporated in analysis of the trial data, when taking a Bayesian approach. The methods proposed enable available ICC information to be summarised appropriately by an informative prior distribution, which is of direct practical use in cluster randomized trials.
Bayes Theorem↗
Reported use of data monitoring committees in the main published reports of randomized controlled trials: a cross-sectional study.
BACKGROUND: We describe a review of published main reports of randomized controlled trials (RCTs), in order to measure the frequency of reported use of data monitoring committees (DMCs) and factors associated with reported DMC use. METHODS: Twenty-four higher impact general and specialist medical journals were handsearched for main reports of RCTs in order to provide a cross-sectional sample of trials published in the year 2000. Additionally, the same general medical journals were handsearched for 1990 to allow a comparison across time. RESULTS: Of 662 RCTs published in 2000, 120 (18%) explicitly reported using a DMC, while 107 (16%) reported planned interim analyses. Overall, about a quarter (24%) reported at least one of these. A higher proportion of trials reported using a DMC in 2000 than 1 990 (70/282, 25% versus 21/204, 10%) in the general medical journals. Logistic regression models suggested the more important variables associated with increased reported DMC use were: later year of publication, publication in general medical journal, survival-based endpoint, multicentre trial, increasing number of patients recruited, at least one arm involving a placebo, at least one arm involving a drug, factorial design and USA involvement in the trial. CONCLUSIONS: In 2000, about a quarter of main RCT reports mention use of a DMC. Actual use of DMCs is likely to be somewhat greater. Reporting use of a DMC was more likely for larger and longer trials among other factors. We believe the factors affecting reported use affect actual use. It is recommended that when a DMC oversees a trial, brief details should be explicitly included in the main trial paper. Standard nomenclature for DMCs is recommended.
Clinical Trials Data Monitoring Committees↗
Systematic qualitative review of the literature on data monitoring committees for randomized controlled trials.
AIMS: To systematically review the published literature on data monitoring committees (DMCs) for randomized controlled trials (RCT) and summarize information and opinions on best practice. This was part of the DAMOCLES project. METHODS: A systematic and comprehensive search of five online bibliographic databases was performed, identifying 4007 potentially relevant articles. These were assessed in two stages by the authors. The 84 most relevant articles were agreed and were supplemented with extracts from 16 books: ultimately, 100 sources were reviewed. A series of 23 questions plus subquestions were developed to structure the data extraction and interpretation process. RESULTS: Much has been written about DMCs but by a rather small community of authors. The papers included some results of surveys, but were mainly opinion pieces based on the authors' beliefs, practices and experiences. There is a lack of empirical evidence for many aspects of DMCs. There was a great range of detail in the literature relating to the prespecified questions. It was generally agreed that interim monitoring of accumulating data was necessary in some form for most trials. Questions such as membership of the DMC featured widely in the literature with opinions and practice ranging from 3-20 members, of whom between none and all should be independent. There was a consensus that formal statistical methods should be used as tools to guide decision making rather than as hard rules. Conversely, topics such as the training and experience required for DMC membership were discussed in very few papers. CONCLUSIONS: There is a consensus in the published literature in a number of areas, although there are many different models for structure and functions of a DMC. While uncertainty remains about some issues, it is strongly recommended that an explicit set of guidelines (Charter) is prepared for each DMC prior to the start of the trial specifying clearly how it will operate.
Clinical Trials Data Monitoring Committees↗
Bayesian approaches to multiple sources of evidence and uncertainty in complex cost-effectiveness modelling.
Increasingly complex models are being used to evaluate the cost-effectiveness of medical interventions. We describe the multiple sources of uncertainty that are relevant to such models, and their relation to either probabilistic or deterministic sensitivity analysis. A Bayesian approach appears natural in this context. We explore how sensitivity analysis to patient heterogeneity and parameter uncertainty can be simultaneously investigated, and illustrate the necessary computation when expected costs and benefits can be calculated in closed form, such as in discrete-time discrete-state Markov models. Information about parameters can either be expressed as a prior distribution, or derived as a posterior distribution given a generalized synthesis of available data in which multiple sources of evidence can be differentially weighted according to their assumed quality. The resulting joint posterior distributions on costs and benefits can then provide inferences on incremental cost-effectiveness, best presented as posterior distributions over net-benefit and cost-effectiveness acceptability curves. These ideas are illustrated with a detailed running example concerning the cost-effectiveness of hip prostheses in different age-sex subgroups. All computations are carried out using freely available software for conducting Markov chain Monte Carlo analysis.
Adult↗
Mortality and volume of cases in paediatric cardiac surgery: retrospective study based on routinely collected data.
OBJECTIVES: To determine whether mortality between 1991 and 1995 in hospitals in England carrying out surgery for congenital heart disease in children was associated with the annual volume of cases and to estimate the extent to which an association could explain the apparent divergent mortality at Bristol Royal Infirmary. DESIGN: Retrospective analysis of data from two sources, a register of returns by surgeons to their professional society and an administrative database. SETTING: 12 hospitals in England carrying out surgery for congenital heart disease over the period April 1991 to March 1995. MAIN OUTCOME MEASURE: 30 day mortality. RESULTS: For open heart operations in children under 1 year old, and in particular for arterial switches and repair of atrioventricular septal defect, there is strong and consistent evidence of an inverse association between mortality and volume of cases (not taking into account any data from Bristol). A hospital carrying out 120 open operations per year in 1991-5 on children aged under 1 year would be expected to have a mortality 25% lower than that in a hospital carrying out 40 operations. If the children in the hospitals had the same mix of operations, this reduction is 34%. Stratifying for types of operation or including the results from Bristol strengthens this association. It was also estimated that less than a fifth of the excess mortality at Bristol Royal Infirmary in open operations in children less than 1 year old was due to the hospital's lower volume of surgery. CONCLUSIONS: Using appropriate methods, this study showed that mortality in paediatric cardiac surgery was inversely related to the volume of surgery. Considerable caution is needed in interpreting these results, and it does not necessarily follow that concentrating resources in fewer centres would reduce mortality.
Cardiology Service, Hospital↗
Being sceptical about meta-analyses: a Bayesian perspective on magnesium trials in myocardial infarction.
BACKGROUND: There has been extensive discussion of the apparent conflict between meta-analyses and a mega-trial investigating the benefits of intravenous magnesium following myocardial infarction, in which the early trial results have been said to be 'too good to be true'. METHODS: We apply Bayesian methods of meta-analysis to the trials available before and after the publication of the ISIS-4 results. We show how scepticism can be formally incorporated into an analysis as a Bayesian prior distribution, and how Bayesian meta-analysis models allow appropriate exploration of hypotheses that the treatment effect depends on the size of the trial or the risk in the control group. RESULTS: Adoption of a sceptical prior would have led early enthusiasm for magnesium to be suitably tempered, but only if combined with a random effects meta-analysis, rather than the fixed effect analysis that was actually conducted. CONCLUSIONS: We argue that neither a fixed effect nor a random effects analysis is appropriate when the mega-trial is included. The Bayesian framework provides many possibilities for flexible exploration of clinical hypotheses, but there can be considerable sensitivity to apparently innocuous assumptions.
Bayes Theorem↗