SEARCH · PubMed Health

Results for “Bayesian analysis”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,117 records · Page 62Linked to original sources

PubMed2005

Intravenous busulfan in adults prior to haematopoietic stem cell transplantation: a population pharmacokinetic study.

An IV form of busulfan (IV Bu) has recently become available for high dose conditioning regimen before haematopoietic stem cell transplantation (HSCT). This IV form is expected to reduce the high pharmacokinetic variability exhibited with oral busulfan and as a result, to better target the plasma area under the curve (AUC). Pharmacokinetics (PK) of IV Bu was investigated on 127 adult patients (333 PK administrations) who received 0.8 mg.kg-1 of Bu as a 2-h infusion every 6 h over 4 days, followed by cyclophosphamide (60 mg.kg-1 day-1x2). A retrospective population PK analysis was carried out to search for important predictive factors of IV Bu PK and to develop a limited sampling strategy (LSS) through Bayesian methodology. The analysis was conducted using the Non Linear Mixed Effect methodology and included a validation process on an independent data set. Adjusted Ideal Body Weight (AIBW) and Body Surface Area (BSA) were the best covariates to explain the inter-patient variability. The final inter-patient variability (CV=16%) in IV Bu clearance (Cltot) was estimated close to the intra-patient variability (CV=13%). There was neither age-dependency nor gender effect. IV Bu Cltot was not affected by elevated hepatic enzymes or by co-administration of either fluconazole or acetaminophen, and was not altered in heavily pre-treated or pre-transplanted patients. Normalised Cltot based on either AIBW or BSA was comparable between normal and obese patients (BMI=18-26.9 kg.m-2, >26.9 kg.m-2, respectively) whereas significant differences existed when based on either actual (ABW) or ideal body weight (IBW). As a consequence, no dose adjustment is required in obese patients when using a AIBW- or BSA-based dose calculation. A fixed dose of 0.80 mg.kg-1 of AIBW or 29 mg.m-2 of BSA yielded an average AUC of 1,200 microM.min, with 80% of patients within the "therapeutic" AUC range of 900-1,500 microM.min. Alternatively, 0.80 mg.kg-1 based on either ABW or IBW for normal patients and on AIBW for obese patients would achieve the same performance. A limited sampling strategy based on a Bayesian methodology was developed and validated on an independent dataset: AUCs obtained from one to two samplings were demonstrated to be reliably estimated.

Adult↗

PubMed2001

Application of a Bayesian method for optimal subset regression to linkage analysis of Q1 and Q2.

We explore an approach that allows us to consider a trait for which we wish to determine the optimal subset of markers out of a set of p > or = 3 candidate markers being considered in a linkage analysis. The most effective analysis would find the model that only includes the q markers closest to the q major genes which determine the trait. Finding this optimal model using classical "frequentist" multiple regression techniques would require consideration of all 2p possible subsets. We apply the work of George and McCulloch [J Am Stat Assoc 88:881-9, 1993], who have developed a Bayesian approach to optimal subset selection regression, to a modification of the Haseman-Elston linkage statistic [Elston et al., Genet Epidemiol 19:1-17, 2000] in the analysis of the two quantitative traits simulated in Problem 2. The results obtained using this Bayesian method are compared to those obtained using (1) multiple regression and (2) the modified Haseman-Elston method (single variable regression analysis). We note upon doing this that for both Q1 and Q2, (1) we have extremely low power with all methods using the samples as given and have to resort to combining several simulated samples in order to have power of 50%, (2) the multivariate analysis does not have greater power than the univariate analysis for these traits, and (3) the Bayesian approach identifies the correct model more frequently than the frequentist approaches but shows no clear advantage over the multivariate approach.

Bayes Theorem↗

PubMed2006

Curious phenomena in Bayesian adjustment for exposure misclassification.

Many epidemiologic investigations involve some discussion of exposure misclassification, but rarely is there an attempt to adjust for misclassification formally in the statistical analysis. Rather, investigators tend to rely on intuition to comment qualitatively on how misclassification might impact their findings. We point out several ways in which intuition might fail, in the context of unmatched case-control analysis with non-differential exposure misclassification. Particularly, we focus on how intuition can conflict with the results of a Bayesian analysis that accounts for the various uncertainties at hand. First, the Bayesian adjustment for misclassification can weaken the evidence about the direction of an exposure-disease association. Second, admitting uncertainty about the misclassification parameters can lead to narrower interval estimates concerning the association. We focus on the simple setting of unmatched case-control analysis with binary exposure and without adjustment for confounders, though much of our discussion should be relevant more generally.

Anti-Bacterial Agents↗

PubMed1978

Multivariate approach to predictive diagnosis of bone-marrow iron stores.

The authors constructed a computer-assisted multivariate data analysis system that finds the subset of a group of laboratory tests (i.e., experimental variables) best able to discriminate between previously identified subpopulations. A Bayesian decision-analysis algorithm was devised for predictive diagnosis by the use of sample means and standard deviations of those identified tests. This approach was used on a set of nine hematologic laboratory tests to find a subset capable of efficiently predicting the presence or absence of bone-marrow iron stores. This identified subset proved to have a diagnostic efficiency of 0.90. The success of this system suggests the feasibility of such an approach in finding a "best" subset of loosely related variables.

Anemia, Hypochromic↗

PubMedDate not supplied

An iterative Bayesian approach to health technology assessment: application to a policy of preoperative optimization for patients undergoing major elective surgery.

PURPOSE: This article presents an iterative framework for managing the dynamic process of health technology assessment. The framework uses Bayesian statistical decision theory and value of information (VOI) analysis to inform decision making regarding appropriate patient management and to direct future research effort over the lifetime of a technology. Within the article, the framework is applied to a policy decision regarding preoperative patient management before major elective surgery, for which trial data are available. METHOD: The evidence available prior to the trial is used to determine the appropriate method of patient management and to ascertain whether, at the time of commissioning, the trial was potentially worthwhile. The prior information is then updated with the trial data via a Bayesian analysis using informative priors. This post trial information set is then used to reassess the appropriate method for patient management and to determine whether there is a requirement for any further research. RESULTS: Prior to the trial, preoperative optimization with dopexamine is identified as the appropriate method of patient management. The results of the VOI analysis suggest that a short-term trial was potentially worthwhile (population expected value of perfect information [EVPI] = 48 million pounds sterling). Following the trial, the uncertainty surrounding the choice of appropriate patient management and the potential worth of further research had increased (population EVPI = 67 million pounds sterling). CONCLUSIONS: The article demonstrates the value and practicality of applying the iterative framework to the dynamic process of health technology assessment. It is only by formally incorporating all of the information available to decision makers, through informed priors, that the appropriate decisions can be made.

Aged↗

PubMed2003

Phylogeny and classification of the Digenea (Platyhelminthes: Trematoda).

Complete small subunit ribosomal RNA gene (ssrDNA) and partial (D1-D3) large subunit ribosomal RNA gene (lsrDNA) sequences were used to estimate the phylogeny of the Digenea via maximum parsimony and Bayesian inference. Here we contribute 80 new ssrDNA and 124 new lsrDNA sequences. Fully complementary data sets of the two genes were assembled from newly generated and previously published sequences and comprised 163 digenean taxa representing 77 nominal families and seven aspidogastrean outgroup taxa representing three families. Analyses were conducted on the genes independently as well as combined and separate analyses including only the higher plagiorchiidan taxa were performed using a reduced-taxon alignment including additional characters that could not be otherwise unambiguously aligned. The combined data analyses yielded the most strongly supported results and differences between the two methods of analysis were primarily in their degree of resolution. The Bayesian analysis including all taxa and characters, and incorporating a model of nucleotide substitution (general-time-reversible with among-site rate heterogeneity), was considered the best estimate of the phylogeny and was used to evaluate their classification and evolution. In broad terms, the Digenea forms a dichotomy that is split between a lineage leading to the Brachylaimoidea, Diplostomoidea and Schistosomatoidea (collectively the Diplostomida nomen novum (nom. nov.)) and the remainder of the Digenea (the Plagiorchiida), in which the Bivesiculata nom. nov. and Transversotremata nom. nov. form the two most basal lineages, followed by the Hemiurata. The remainder of the Plagiorchiida forms a large number of independent lineages leading to the crown clade Xiphidiata nom. nov. that comprises the Allocreadioidea, Gorgoderoidea, Microphalloidea and Plagiorchioidea, which are united by the presence of a penetrating stylet in their cercariae. Although a majority of families and to a lesser degree, superfamilies are supported as currently defined, the traditional divisions of the Echinostomida, Plagiorchiida and Strigeida were found to comprise non-natural assemblages. Therefore, the membership of established higher taxa are emended, new taxa erected and a revised, phylogenetically based classification proposed and discussed in light of ontogeny, morphology and taxonomic history.

Animals↗

PubMed2005

[Lung cancer mortality among women in France. Trend analysis and projection between 1975 and 2014, with a bayesian age-cohort model].

BACKGROUND: With 4,500 deaths in year 2000, female lung cancer mortality rates increased by 3% every year over the last two decades in France. This trend, not observed among males, is attributed to the regular increase of female smoking. In order to answer French Health decider's concerns, we estimated the future female lung cancer mortality rates and numbers of deaths for the next fifteen years, in France and its regions. METHODS: Analyses were based on numbers of female deaths from lung cancer observed between 1975 and 1999, and on past and future population estimates for 1975-2014, at national and regional levels. Mortality rates and numbers of deaths in France and its regions by 5-year periods and 5-year age groups were given in the 1975-1999 death certificate data base, and were projected for 2000-2014. The analysis used a bayesian approach of the age-cohort model with auto-regressive constraints on parameters. Estimated mortality rates were standardized on truncated 20-85 + world population. RESULTS: French female lung cancer mortality increased by 3% every year between 1975 and 1999. In period 1995-1999, truncated 20-85 + mortality rates, and number of deaths per year were respectively 11.4 per 100,000 and 4,000. Mortality rates increased in all regions but variations were maximum in Corsica (+ 314%) and minimum in Auvergne (+ 37%). For the whole of France, the estimated truncated 20-85 + standardized rate, was respectively, 14.1 and 22.5 per 100,000 in period 2000-04 and period 2010-14, which represents a 60% increase between these two time periods. At the regional level, the maximum variation was found in Languedoc-Roussillon (107%), the minimum in Nord-Pas-de-Calais (40%). CONCLUSIONS: The bayesian approach of the age-cohort model is increasingly used because it produces stable projections, without having to include other cancer parameters. Nevertheless, it would be interesting to extend this model by incorporating a tobacco consumption component, in order to assess scenarios based on consumption decreases.

Adult↗

PubMed1998

Population analysis of once-daily dosing of gentamicin in patients with neutropenia.

BACKGROUND: Aminoglycoside antibiotics are used as first-line therapy in the treatment of patients with febrile neutropenia. However, there is little information on the pharmacokinetics of aminoglycosides in this specific patient group and whether they differ from the accepted population values. AIM: To determine the population pharmacokinetics of gentamicin for patients with febrile neutropenia, with or without sepsis. METHODS: Data were collected from 26 patients with febrile neutropenia receiving once-daily gentamicin. Patient age, height, weight, estimated lean body weight (LBW), gender, serum creatinine (SCr), creatinine clearance (Cl(Cr)), and serum drug concentration data were collected. Severity of infection and degree of neutropenia were assessed. An initial two-stage population analysis using a Bayesian dose-individualisation programme was performed to estimate likely population values of the parameters. If these were significantly different from typical values then a true population analysis was to be performed. RESULTS: Data for 13 female and 13 male patients were collected. Median age was 57 years (range 19 to 87 years), estimated LBW (kg) = 66 (+/- 10), Cl(Cr) (1/hour) = 5.6 (+/- 2.0) (mean +/- SD). Results of the two-stage approach follow [initial population values in square brackets]: Cl (1/hour) = 0.73 (+/- 0.12) x Cl(Cr) + 0.01 x LBW [0.70 (+/- 0.27) x Cl(Cr) + 0.01 x LBW]; Vd (1) = 0.28 (+/- 0.05) x LBW [0.27 (+/- 0.06) x LBW]. CONCLUSIONS: The population values of Cl and Vd from the two-stage approach do not differ significantly from initial population values of Cl and Vd. Patients receiving gentamicin for febrile neutropenia may be considered pharmacokinetically similar to the general population receiving gentamicin.

Adult↗

PubMed2006

Missing value estimation for DNA microarray gene expression data by Support Vector Regression imputation and orthogonal coding scheme.

BACKGROUND: Gene expression profiling has become a useful biological resource in recent years, and it plays an important role in a broad range of areas in biology. The raw gene expression data, usually in the form of large matrix, may contain missing values. The downstream analysis methods that postulate complete matrix input are thus not applicable. Several methods have been developed to solve this problem, such as K nearest neighbor impute method, Bayesian principal components analysis impute method, etc. In this paper, we introduce a novel imputing approach based on the Support Vector Regression (SVR) method. The proposed approach utilizes an orthogonal coding input scheme, which makes use of multi-missing values in one row of a certain gene expression profile and imputes the missing value into a much higher dimensional space, to obtain better performance. RESULTS: A comparative study of our method with the previously developed methods has been presented for the estimation of the missing values on six gene expression data sets. Among the three different input-vector coding schemes we tried, the orthogonal input coding scheme obtains the best estimation results with the minimum Normalized Root Mean Squared Error (NRMSE). The results also demonstrate that the SVR method has powerful estimation ability on different kinds of data sets with relatively small NRMSE. CONCLUSION: The SVR impute method shows better performance than, or at least comparable with, the previously developed methods in present research. The outstanding estimation ability of this impute method is partly due to the use of the most missing value information by incorporating orthogonal input coding scheme. In addition, the solid theoretical foundation of SVR method also helps in estimation of performance together with orthogonal input coding scheme. The promising estimation ability demonstrated in the results section suggests that the proposed approach provides a proper solution to the missing value estimation problem. The source code of the SVR method is available from http://202.38.78.189/downloads/svrimpute.html for non-commercial use.

Algorithms↗

PubMed2000

Risk of HIV infection as a function of the duration of intravenous drug use: a non-parametric Bayesian approach.

We analyse the elapsed time between intravenous (IV) drug initiation and HIV infection in a cohort of 972 injecting drug users attending a hospital detoxification unit. We use the time of seroconversion instead of the time of HIV infection because the date of HIV infection is rarely known and the gap between these two times is negligible (around one to three months). Although seroconversion time cannot be determined exactly, it can be inferred at least to within an interval. This seroconversion interval is determined from the dates of HIV antibody tests, if available. The data is consequently interval-censored. We estimate the distribution function of the elapsed time from IV drug initiation to seroconversion as well as the risk of seroconversion by means of a non-parametric Bayesian approach. The analysis is conducted according to the following four calendar periods: before or at 1980; between 1981 and 1985; between 1986 and 1991; after or at 1992 where the IV drug use was initiated. The methodology used is based on an alternating conditional sampling algorithm. The Bayesian approach allows not only the incorporation of prior beliefs about the distribution function, but also the analysis of the risk of seroconversion without assuming restrictive parametric models. Furthermore, the estimator for the distribution function is smooth and thus differences between groups can be easily interpreted.

Adolescent↗

PubMed1996

Bayesian inference on prevalence using a missing-data approach with simulation-based techniques: applications to HIV screening.

Health departments and other health-related authorities seek accurate assessment of the spread of human immunodeficiency virus (HIV) among populations. Although screening for HIV provides a direct means for estimating its prevalence, it is complicated by the heterogeneity of available diagnostic tests and the degree to which they can diagnose HIV accurately. To integrate the limited precision of screening tests with prior results, Bayesian inference becomes a method of choice. Current Bayesian methods, however, have limited applications and do not readily generalize for complicated sampling designs and for modelling needs, particularly those that relate to HIV screening. By utilizing recent developments in the theories of missing-data analysis and simulation-based techniques, we develop an approach to Bayesian analysis of prevalence. This methodology is quite general for a variety of sampling schemes and sufficiently flexible to accommodate various practical considerations that arise from HIV screening. We illustrate the methodology with real as well as simulated data sets. Further, by utilizing the methodology, we performed simulations to demonstrate that pooled testing provides a cost-effective means to improve the precision of estimates of prevalence under the currently limited screening technology.

Algorithms↗

PubMed2006

BNTagger: improved tagging SNP selection using Bayesian networks.

Genetic variation analysis holds much promise as a basis for disease-gene association. However, due to the tremendous number of candidate single nucleotide polymorphisms (SNPs), there is a clear need to expedite genotyping by selecting and considering only a subset of all SNPs. This process is known as tagging SNP selection. Several methods for tagging SNP selection have been proposed, and have shown promising results. However, most of them rely on strong assumptions such as prior block-partitioning, bi-allelic SNPs, or a fixed number or location of tagging SNPs. We introduce BNTagger, a new method for tagging SNP selection, based on conditional independence among SNPs. Using the formalism of Bayesian networks (BNs), our system aims to select a subset of independent and highly predictive SNPs. Similar to previous prediction-based methods, we aim to maximize the prediction accuracy of tagging SNPs, but unlike them, we neither fix the number nor the location of predictive tagging SNPs, nor require SNPs to be bi-allelic. In addition, for newly-genotyped samples, BNTagger directly uses genotype data as input, while producing as output haplotype data of all SNPs. Using three public data sets, we compare the prediction performance of our method to that of three state-of-the-art tagging SNP selection methods. The results demonstrate that our method consistently improves upon previous methods in terms of prediction accuracy. Moreover, our method retains its good performance even when a very small number of tagging SNPs are used.

Algorithms↗

PubMed1994

Nonparametric estimation of population characteristics of the kinetics of lithium from observational and experimental data: individualization of chronic dosing regimen using a new Bayesian approach.

A population analysis of the kinetics of lithium was performed from experimental and observational data in 113 subjects in order to propose a new approach for lithium dosage individualization. The kinetics of lithium is described by a two-compartment model. Age, body weight, height, and serum creatinine are included as covariates. Population analysis was performed by the nonparametric maximum likelihood method, which provides an estimate of the distribution of the five kinetic parameters and covariates. Mean lithium clearance was 1.50 L/h with a coefficient of variation (CV) of 38%, and was found to increase with body weight. Results were consistent with those of earlier studies and confirm a large interindividual variability. Data from a separate group of 35 patients were used to validate results: the estimated a priori and on covariate conditional distributions of the measured 24-h serum lithium concentration following a single dose were consistent with the corresponding measurements. A Bayesian approach for individualizing dosing schemes is proposed. This approach is based on minimization of a risk function expressing the deviation of the trough concentration at steady-state from the therapeutic range.

Adolescent↗

PubMed2005

Investigating heterogeneity in studies of resting energy expenditure in persons with HIV/AIDS: a meta-analysis.

BACKGROUND: There is conflict in the literature about the extent of alterations of resting energy expenditure (REE) in persons with HIV. OBJECTIVE: The study was conducted to ascertain the mean difference in REE (in kJ) per kilogram of fat-free mass (FFM; REE/FFM) between HIV-positive subjects and control subjects and to investigate heterogeneity in the literature. DESIGN: A meta-analysis comparing classical and Bayesian methods was conducted. Heterogeneity was investigated by using subgroup analysis, metaregression, and a mixed indirect comparison. RESULTS: Of 58 studies meeting the inclusion criteria, 32 included both HIV-positive and control groups; 24 of these 32 were included. Thirty-seven studies were used in the mixed indirect comparison, and 30 were used in the subgroup comparisons of the HIV-symptomatic, lipodystrophy, weight-losing, and weight-stable subgroups and the healthy (HIV-negative) control group. Mean REE/FFM was significantly higher in 732 HIV-positive subjects than in 340 control subjects [11.93 kJ/kg (95% CI: 8.44,15.43 kJ/kg) and 12.47 kJ/kg (95% CI: 8.19,16.57 kJ/kg), classical and Bayesian random effects, respectively]; the test for heterogeneity was significant (P < 0.001). Both the mixed indirect comparison and the subgroup analysis indicated that REE/FFM was highest in the symptomatic subgroup; however, the small number of studies investigating symptomatic subjects limited statistical comparisons. The presence of lipodystrophy, use of highly active antiretroviral therapy, subject age, and method of body-composition measurement could not explain the heterogeneity in the data with the use of metaregression. CONCLUSIONS: REE/FFM (kJ/kg) is significantly higher in HIV-positive subjects than in healthy control subjects. Symptomatic HIV infection may contribute to the variations reported in the literature.

Acquired Immunodeficiency Syndrome↗

PubMed2003

Significance of micro-geographical population structure in forensic cases: a bayesian exploration.

We studied the influence of population structure at the microgeographical level on the analysis of forensic cases. A total of nine autosomal STRs and seven Y-STRs were analyzed in the general mixed population and in two relatively isolated valleys of Cantabria, a region in Northern Spain. Statistically significant differences existed in the frequency distribution of four autosomal STRs, with an overall Fst value of 0.3%. A simulation of virtual trio cases revealed that it did not have a practical influence on the analysis of paternity disputes. Significant differences also existed in most Y-STRs, with an overall Fst value of 3%. Thus, using the general database instead of the specific valley database resulted in 5-fold or higher overestimation of the likelihood ratio of matching in up to 30% of cases. A bayesian analysis revealed that this had a significant impact on the estimation of the probability of identity in scenarios of low "a priori" odds of suspicion.

Alleles↗

PubMed2005

A Bayesian threshold-normal mixture model for analysis of a continuous mastitis-related trait.

Mastitis is associated with elevated somatic cell count in milk, inducing a positive correlation between milk somatic cell score (SCS) and the absence or presence of the disease. In most countries, selection against mastitis has focused on selecting parents with genetic evaluations that have low SCS. Univariate or multivariate mixed linear models have been used for statistical description of SCS. However, an observation of SCS can be regarded as drawn from a 2- (or more) component mixture defined by the (usually) unknown health status of a cow at the test-day on which SCS is recorded. A hierarchical 2-component mixture model was developed, assuming that the health status affecting the recorded test-day SCS is completely specified by an underlying liability variable. Based on the observed SCS, inferences can be drawn about disease status and parameters of both SCS and liability to mastitis. The prior probability of putative mastitis was allowed to vary between subgroups (e.g., herds, families), by specifying fixed and random effects affecting both SCS and liability. Using simulation, it was found that a Bayesian model fitted to the data yielded parameter estimates close to their true values. The model provides selection criteria that are more appealing than selection for lower SCS. The proposed model can be extended to handle a wide range of problems related to genetic analyses of mixture traits.

Animals↗

PubMed2001

Bayesian estimation of fold-changes in the analysis of gene expression: the PFOLD algorithm.

A general and detailed noise model for the DNA microarray measurement of gene expression is presented and used to derive a Bayesian estimation scheme for expression ratios, implemented in a program called PFOLD, which provides not only an estimate of the fold-change in gene expression, but also confidence limits for the change and a P-value quantifying the significance of the change. Although the focus is on oligonucleotide microarray technologies, the scheme can also be applied to cDNA based technologies if parameters for the noise model are provided. The model unifies estimation for all signals in that it provides a seamless transition from very low to very high signal-to-noise ratios, an essential feature for current microarray technologies for which the median signal-to-noise ratios are always moderate. The dual use, as decision statistics in a two-dimensional space, of the P-value and the fold-change is shown to be effective in the ubiquitous problem of detecting changing genes against a background of unchanging genes, leading to markedly higher sensitivities, at equal selectivity, than detection and selection based on the fold-change alone, a current practice until now.

Algorithms↗

PubMed2006

Longitudinal and repeated cross-sectional cluster-randomization designs using mixed effects regression for binary outcomes: bias and coverage of frequentist and Bayesian methods.

As medical applications for cluster randomization designs become more common, investigators look for guidance on optimal methods for estimating the effect of group-based interventions over time. This study examines two distinct cluster randomization designs: (1) the repeated cross-sectional design in which centres are followed over time but patients change, and (2) the longitudinal design in which individual patients are followed over time within treatment clusters. Simulations of each study design stipulated a multiplicative treatment effect (on the log odds scale), between 5 and 15 clusters in each of two treatment arms, and followed over two time periods. Estimation options included linear mixed effects models using restricted maximum likelihood (REML), generalized estimating equations (GEE), mixed effects logistic regression using both penalized quasi likelihood (PQL) and numerical integration, and Bayesian Monte Carlo analysis. For the repeated cross-sectional designs, most methods performed well in terms of bias and coverage when clusters were numerous (30) and variability across clusters of baseline risk and treatment effect was modest. With few clusters (two groups of five) and higher variability, only the Bayesian methods maintained coverage. In the longitudinal designs, the common methods of REML, GEE, or PQL performed poorly when compared to numerical integration, while Bayesian methods demonstrated less bias and better coverage for estimates of both log odds ratios and risk differences. The performance of common statistical tools for the analysis of cluster randomization designs depends heavily on the precise design, the number of clusters, and the variability of baseline outcomes and treatment effects across centres.

Bayes Theorem↗