Yinghui Wei

Predictive distributions for between-study heterogeneity and simple methods for their application in Bayesian meta-analysis.

Numerous meta-analyses in healthcare research combine results from only a small number of studies, for which the variance representing between-study heterogeneity is estimated imprecisely. A Bayesian approach to estimation allows external evidence on the expected magnitude of heterogeneity to be incorporated. The aim of this paper is to provide tools that improve the accessibility of Bayesian meta-analysis. We present two methods for implementing Bayesian meta-analysis, using numerical integration and importance sampling techniques. Based on 14 886 binary outcome meta-analyses in the Cochrane Database of Systematic Reviews, we derive a novel set of predictive distributions for the degree of heterogeneity expected in 80 settings depending on the outcomes assessed and comparisons made. These can be used as prior distributions for heterogeneity in future meta-analyses. The two methods are implemented in R, for which code is provided. Both methods produce equivalent results to standard but more complex Markov chain Monte Carlo approaches. The priors are derived as log-normal distributions for the between-study variance, applicable to meta-analyses of binary outcomes on the log odds-ratio scale. The methods are applied to two example meta-analyses, incorporating the relevant predictive distributions as prior distributions for between-study heterogeneity. We have provided resources to facilitate Bayesian meta-analysis, in a form accessible to applied researchers, which allow relevant prior information on the degree of heterogeneity to be incorporated.

Estimating within-study covariances in multivariate meta-analysis with multiple outcomes.

Multivariate meta-analysis allows the joint synthesis of effect estimates based on multiple outcomes from multiple studies, accounting for the potential correlations among them. However, standard methods for multivariate meta-analysis for multiple outcomes are restricted to problems where the within-study correlation is known or where individual participant data are available. This paper proposes an approach to approximating the within-study covariances based on information about likely correlations between underlying outcomes. We developed methods for both continuous and dichotomous data and for combinations of the two types. An application to a meta-analysis of treatments for stroke illustrates the use of the approximated covariance in multivariate meta-analysis with correlated outcomes. Copyright

Non-invasive prenatal diagnostic test accuracy for fetal sex using cell-free DNA a review and meta-analysis

BackgroundCell-free fetal DNA (cffDNA) can be detected in maternal blood during pregnancy, opening the possibility of early non-invasive prenatal diagnosis for a variety of genetic conditions. Since 1997, many studies have examined the accuracy of prenatal fetal sex determination using cffDNA, particularly for pregnancies at risk of an X-linked condition. Here we report a review and meta-analysis of the published literature to evaluate the use of cffDNA for prenatal determination (diagnosis) of fetal sex. We applied a sensitive search of multiple bibliographic databases including PubMed (MEDLINE), EMBASE, the Cochrane library and Web of Science.ResultsNinety studies, incorporating 9,965 pregnancies and 10,587 fetal sex results met our inclusion criteria. Overall mean sensitivity was 96.6% (95% credible interval 95.2% to 97.7%) and mean specificity was 98.9% (95% CI = 98.1% to 99.4%). These results vary very little with trimester or week of testing, indicating that the performance of the test is reliably high.ConclusionsBased on this review and meta-analysis we conclude that fetal sex can be determined with a high level of accuracy by analyzing cffDNA. Using cffDNA in prenatal diagnosis to replace or complement existing invasive methods can remove or reduce the risk of miscarriage. Future work should concentrate on the economic and ethical considerations of implementing an early non-invasive test for fetal sex.

Bayesian multivariate meta-analysis with multiple outcomes

There has been a recent growth in developments of multivariate meta-analysis. We extend the methodology of Bayesian multivariate meta-analysis to the situation when there are more than two outcomes of interest, which is underexplored in the current literature. Our objective is to meta-analyse summary data from multiple outcomes simultaneously, accounting for potential dependencies among the data. One common issue is that studies do not all report all of the outcomes of interests, and we take an approach relying on marginal modelling of only the reported data. We employ a separation prior for the between-study variance-covariance matrix, which offers an improvement on the conventional inverse-Wishart prior, showing robustness in estimation and flexibility in incorporating prior information. Particular challenges arise when the number of outcomes is large relative to the number of studies because the number of parameters in the variance-covariance matrix can become substantial and there can be very little information with which to estimate between-study correlation coefficients. We explore assumptions that reduce the number of parameters in this matrix, including assumptions of homogenous variances, homogenous correlations for certain outcomes and positive correlation coefficients. We illustrate the methods with an example data set from the Cochrane Database of Systematic Reviews.

Comparative efficacy and safety of treatments for localised prostate cancer: an application of network meta-analysis

Context There is ongoing uncertainty about the optimal management of patients with localised prostate cancer. Objective To evaluate the comparative efficacy and safety of different treatments for patients with localised prostate cancer. Design Systematic review with Bayesian network meta-analysis to estimate comparative ORs, and a score (0–100%) that, for a given outcome, reflects average rank order of superiority of each treatment compared against all others, using the Surface Under the Cumulative RAnking curve (SUCRA) statistic. Data sources Electronic searches of MEDLINE without language restriction. Study selection Randomised trials comparing the efficacy and safety of different primary treatments (48 papers from 21 randomised trials included 7350 men). Data extraction 2 reviewers independently extracted data and assessed risk of bias. Results Comparative efficacy and safety evidence was available for prostatectomy, external beam radiotherapy (different types and regimens), observational management and cryotherapy, but not high-intensity focused ultrasound. There was no evidence of superiority for any of the compared treatments in respect of all-cause mortality after 5 years. Cryotherapy was associated with less gastrointestinal and genitourinary toxicity than radiotherapy (SUCRA: 99% and 77% for gastrointestinal and genitourinary toxicity, respectively). Conclusions The limited available evidence suggests that different treatments may be optimal for different efficacy and safety outcomes. These findings highlight the importance of informed patient choice and shared decision-making about treatment modality and acceptable trade-offs between different outcomes. More trial evidence is required to reduce uncertainty. Network meta-analysis may be useful to optimise the power of evidence synthesis studies once data from new randomised controlled studies in this field are published in the future.

Meta-analysis of time-to-event outcomes from randomized trials using restricted mean survival time: application to individual participant data

Meta-analysis of time-to-event outcomes using the hazard ratio as a treatment effect measure has an underlying assumption that hazards are proportional. The between-arm difference in the restricted mean survival time is a measure that avoids this assumption and allows the treatment effect to vary with time. We describe and evaluate meta-analysis based on the restricted mean survival time for dealing with non-proportional hazards and present a diagnostic method for the overall proportional hazards assumption. The methods are illustrated with the application to two individual participant meta-analyses in cancer. The examples were chosen because they differ in disease severity and the patterns of follow-up, in order to understand the potential impacts on the hazards and the overall effect estimates. We further investigate the estimation methods for restricted mean survival time by a simulation study.

Evaluating hospital infection control measures for antimicrobial-resistant pathogens using stochastic transmission models: Application to vancomycin-resistant enterococci in intensive care units

Nosocomial pathogens such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococci (VRE) are the cause of significant morbidity and mortality among hospital patients. It is important to be able to assess the efficacy of control measures using data on patient outcomes. In this paper, we describe methods for analysing such data using patient-level stochastic models which seek to describe the underlying unobserved process of transmission. The methods are applied to detailed longitudinal patient-level data on vancomycin-resistant Enterococci from a study in a US hospital with eight intensive care units (ICUs). The data comprise admission and discharge dates, dates and results of screening tests, and dates during which precautionary measures were in place for each patient during the study period. Results include estimates of the efficacy of the control measures, the proportion of unobserved patients colonized with vancomycin-resistant Enterococci, and the proportion of patients colonized on admission.

Parent–infant psychotherapy: a systematic review of the evidence for improving parental and infant mental health

Abstract Background: Parent–infant psychotherapy (PIP) is a psychodynamic intervention with parent–infant dyads, designed to address regulatory disturbances in infancy and problems in the parent–infant relationship. Aims: This systematic review aimed to examine whether PIP is effective in improving the parent–infant relationship or other aspects of parent or infant functioning. Methods: A systematic review was undertaken. Electronic databases were searched for randomised controlled trials in which participants had been allocated to a PIP intervention or control group/other treatment. Results: Eight studies were identified that provided data comparing parent–infant psychotherapy with a no-treatment control group (four studies) or comparing PIP with other kinds of treatment (four studies). Meta-analyses indicated that parents who received PIP were more likely to have an infant who was rated as being securely attached to the parent after the intervention; however, there were no significant differences in studies comparing outcomes of PIP with another model of treatment. Conclusions: Although PIP appears to be a promising method of improving infant attachment security, there is inconclusive evidence of its benefits in terms of other outcomes, and no evidence to show that it is more effective than other interventions for parents and infants. Many studies had limitations in their design or implementation, and findings must be interpreted with caution.

Physical activity and the prevention, reduction, and treatment of alcohol and/or substance use across the lifespan (The PHASE review): protocol for a systematic review

BackgroundAlcohol and substance use results in significant human and economic cost globally and is associated with economic costs of £21 billion and £15billion within the UK, respectively, and trends for use are not improving. Pharmacological interventions are well researched, but relapse rates across interventions for substance and alcohol use disorders are as high as 60–90%. Physical activity may offer an alternative or adjunct approach to reducing rates of alcohol and substance use that is associated with few adverse side effects, is easily accessible, and is potentially cost-effective. Through psychological, behavioural, and physiological mechanisms, physical activity may offer benefits in the prevention, reduction, and treatment of alcohol and substance use across the lifespan. Whilst physical activity is widely advocated as offering benefit, no systematic review exists of physical activity (in all forms) and its effects on all levels of alcohol and substance use across all ages to help inform policymakers, service providers, and commissioners.MethodsThe objectives of this mixed methods systematic review are to describe and evaluate the quantitative and qualitative research obtained by a diverse search strategy on the impact of physical activity and its potential to:1.Reduce the risk of progression to alcohol and/or substance use (PREVENTION)2.Support individuals to reduce alcohol and/or substance use for harm reduction (REDUCTION), and3.Promote abstinence and relapse prevention during and after treatment for an alcohol and/or substance use disorder (TREATMENT).With the input of key stakeholders, we aim to assess how what we know can be translated into policy and practice. Quantitative, qualitative, service evaluations, and economic analyses will be brought together in a final narrative synthesis that will describe the potential benefits of physical activity for whom, in what conditions, and in what form.DiscussionThis review will provide details of what is known about physical activity and the prevention, reduction, and treatment of alcohol and/or substance use. The synthesised findings will be disseminated to policymakers, service providers, and commissioners in the UK.Systematic review registrationPROSPERO number: CRD42017079322.

Chronic respiratory symptoms and lung abnormalities among people with a history of tuberculosis in Uganda: a national survey.

BACKGROUND People with pulmonary tuberculosis are at risk of developing chronic respiratory disorders due to residual lung damage. To date, the scope of the problem in high-burden tuberculosis countries is relatively unknown. METHODS Chronic respiratory symptoms (cough and phlegm lasting >2 weeks) and radiological lung abnormalities were compared between adults with and without a history of tuberculosis among the general population of Uganda. Multivariable regression models were used to estimate odds ratios (ORs) with adjustment for age, gender, smoking, education, setting, and region. Random effects models accounted for village clustering effect. RESULTS Of 45293 invited people from 70 villages, 41154 (90.9%) participated in the survey. A total of 798 had a history of tuberculosis and, among them, 16% had respiratory symptoms and 41% X-ray abnormalities. Adjusted ORs showed strong evidence for individuals with a history of tuberculosis having increased risk of respiratory symptoms (OR, 4.02; 95% confidence interval [CI], 3.25-4.96) and X-ray abnormalities (OR, 17.52; 95% CI, 14.76-20.79), attributing 6% and 24% of the respective population risks. CONCLUSIONS In Uganda, a history of tuberculosis was a strong predictor of respiratory symptoms and lung abnormalities, before older age and smoking. Eliminating tuberculosis disease could reduce the prevalence of chronic respiratory symptoms as much as eliminating smoking.