Frequentist Approach Research Articles

P1043 Immunomodulator and advanced therapies for prevention of clinical relapse and loss of response during maintenance phase in Crohn’s disease: A systematic review and network meta-analysis

Abstract Background Several effective advanced therapies have been approved for the treatment of Crohn’s disease (CD) in the past two decades. Given lack of adequate head-to head clinical trials it is challenging to position these therapies in the management algorithm. Therefore we performed this network meta-analysis to inform therapeutic decisions. Methods We searched MEDLINE, Embase, and Cochrane Central Register of Controlled Trials between inception and October 2023 for randomised controlled trials (RCTs). Outcomes assessed were clinical relapse (loss of clinical remission), loss of clinical response, adverse events (AE), withdrawal due to AE, and serious AE. We performed random-effects meta-analysis and network meta-analysis (using a frequentist approach), and estimated relative risk (RRs), 95%CI values. We used GRADE to ascertain certainty of evidence. Results A total of 32 RCTs comprising 10,924 patients were included in the analysis. On network meta-analysis of 15 interventions based on prevention of clinical relapse, combination of infliximab (IFX) and azathioprine (AZA) (RR 0.31, 95%CI [0.17-0.56]), IFX (RR 0.62, 95%CI [0.5-0.76], moderate certainty), and adalimumab (RR 0.68, 95%CI [0.55-0.84], moderate certainty) were the only agents that were effective compared to placebo (Figure 1). Methotrexate (RR 0.57, 95%CI [0.32-1.03], low certainty) and azathioprine (RR 0.82, 95%CI [0.62-1.1]), very low certainty) were not effective. On analysis based on prevention of clinical response for nine interventions, there was high certainty evidence for ustekinumab (RR 0.73, 95%CI [0.61-0.86]), upadacitinib (RR 0.64, 95%CI [0.57-0.72]), and moderate certainty of evidence for infliximab (RR 0.71, 95%CI [0.59-0.84]), adalimumab (RR 0.68, 95%CI [0.61-0.75]), certolizumab (RR 0.57, 95%CI [0.47-0.7]), CTP13 (RR 0.46, 95%CI [0.38-0.57]), vedolizumab (RR 0.82, 95%CI [0.68-0.99]), natalizumab (RR 0.54, 95%CI [0.43-0.68]). Tofacitinib was not effective and there was no data available for immunomodulators and combination therapies for this outcome. On analysis of serious adverse events advanced therapies were of no difference to placebo (very low certainty). Conclusion On network meta-analysis immunomodulators and advanced therapies for maintenance phase, combination of anti-TNFs and immunomodulators followed by anti-TNF monotherapy had large effect size with moderate certainty for prevention of clinical relapse. There was moderate to high certainty evidence to suggest newer advanced therapies are effective in prevention of loss of clinical response.

P556 Immunomodulator and advanced therapies for Induction of clinical remission and response in Crohn’s disease: A systematic review and network meta-analysis

Abstract Background Several effective advanced therapies have been approved for the treatment of Crohn’s disease (CD) in the past two decades. Many have used synthesis techniques to study this area, but these rarely consider both advanced and immunomodulator therapy together in the same analysis. Network meta-analysis enables the direct and indirect comparison of these different therapeutic agents, aiding in determining their relative positioning within the treatment algorithm. Methods We searched MEDLINE, Embase, and Cochrane Central Register of Controlled Trials between inception and October 2023 for randomised controlled trials (RCTs). The outcomes included were clinical remission and response, adverse events (AE), withdrawal due to AE, total and serious AE. We performed random-effects meta-analysis and network meta-analysis (frequentist approach), and estimated relative risk (RRs), 95%CI values. We used GRADE to ascertain certainty of evidence. Results A total of 83 RCTs comprising 19214 patients were eligible. A total of 28 interventions were included in the analysis. Combination of anti-tumor necrosis factor (TNF) and immunomodulators, anti-TNF monotherapy (adalimumab [ADA], infliximab [IFX], BI695501), IL-23 antagonists (ustekinumab [UST], guselkumab, risankizumab [RIS], mirikizumab), anti-integrins (vedolizumab, natalizumab) and upadacitinib were effective in inducing clinical remission compared to placebo (Figure 1). There was moderate certainty of evidence for combination of ADA and azathioprine (AZA) (RR 3.1, 95%CI [2.0-4.8]; risk difference [RD]: 40.1%), IFX+AZA (RR 2.5, 95%CI [1.7-3.8]; RD: 28.9%), ADA (RR 2.5, 95%CI [1.8-3.5]; RD:28.5%), and UST (RR 2.0 95%CI, [1.6-2.5]; RD: 19.2%) in induction of clinical remission compared to placebo. For induction of clinical response, there was moderate certainty of evidence for IFX+AZA (RR 2.8, 95%CI [1.6-4.8]), ADA (RR 2.5, 95%CI [1.7-3.9]), IFX (RR 2.8, 95%CI [1.6-4.8]), RIS (RR 1.9, 95%CI [1.3-2.6]). On analysis of serious adverse events all therapies were of no difference to placebo (very low certainty). On sensitivity analysis, the impact of concomitant immunomodulator use and the biological naivety of patients had little effect on these results. Conclusion On network meta-analysis, combination of anti-TNFs and immunomodulators followed by anti-TNF monotherapy had large effect size with moderate certainty for the induction of clinical remission. More novel therapies appear to have examples of similarly important effect sizes, but are currently limited due to the imprecision of the limited evidence base at present and future research should target these therapies in study.

A gentle introduction to network meta-analysis for orthodontists

Network meta-analysis has been widely used to address the limitations of traditional pairwise meta-analysis. Network meta-analysis incorporates all available evidence into a general statistical framework for comparing multiple treatments. The original Bayesian approach offers a statistical framework to address heterogeneity in the evidence and complexity in the data structure when clinical trials with more than two treatment groups are included. Alternative frequentist approaches have been developed and implemented in commonly used statistical software. The aim of this article is to provide a non-technical introduction to the statistical models and assumptions of network meta-analysis, such as consistency and transitivity, for the orthodontics and dental research community. An example was used to demonstrate how to conduct a network meta-analysis and how to use GRADE and CINeMA tools for placing confidence in the NMA effect estimates in a network meta-analysis. The statistical theory behind network meta-analysis is complex, so we strongly encourage close collaboration between orthodontists and experienced statisticians when planning and conducting a network meta-analysis. Network meta-analysis has been proven to be a very useful tool for evidence synthesis because it improves the efficiency of comparative effectiveness research and the quality of decision-making.

A Bayesian approach for evaluating equivalence over multiple groups, and comparison with frequentist tost

Manufacturing and testing of pharmaceutical products frequently occur in multiple facilities within a company’s network. It is of interest to demonstrate equivalence among the alternative testing/manufacturing facilities to ensure product consistency and quality regardless of the facility where it was manufactured/tested. In the Frequentist framework, equivalence testing is well established when comparing two labs or manufacturing facilities; however, when considering more than two labs or production sites, the Frequentist approach may not always offer appropriate or interpretable estimates for demonstrating equivalence among all of them simultaneously. This paper demonstrates the utility of Bayesian methods to the equivalence assessment of multiple groups means, with a comparison against traditional Frequentist methods. We conclude that a Bayesian strategy is very useful for addressing the problem of multi-group equivalence. While it is not our intention to argue that Bayesian methods should always replace Frequentist ones, we show that among the advantages of a Bayesian analysis is that it provides a more nuanced understanding of the degree of similarity among sites than the hypothesis testing underpinning the Frequentist approach.

The COVID generation: Online dyslexia treatment equally effective as face-to-face treatment in a Dutch sample

Due to pandemic-induced lockdown(s) in 2020, dyslexia treatment was forced to move to online platforms. This study examined whether Dutch children who received online treatment progressed as much in their reading and spelling performance as children who received the usual face-to-face treatment. To this end, 254 children who received treatment-as-usual were compared to 162 children who received online treatment with Bayesian methods. The advantage of a Bayesian approach is that it can provide evidence for and against the null hypothesis whereas frequentist approaches only provide evidence against it. We found that children in the online treatment condition received slightly fewer treatment sessions but progressed equally after controlling for the number of sessions compared to the treatment-as-usual condition. These results have clinical and practical implications as they show that reading treatment can be successfully delivered online.

Classical and Bayesian Inference of Unit Gompertz Distribution Based on Progressively Type II Censored Data

In this article, two frequentist approaches and a Bayesian approach employing progressive Type II censored data are used to estimate parameters of a unit Gompertz distribution. In frequentist approach, besides conventional maximum likelihood estimation, maximum product of spacing method is proposed for parameter estimation as an alternative approach to common maximum likelihood method. Both Newton-Raphson and stochastic expectation minimization algorithms are used for computing the MLEs, while Bayes estimates are obtained using both the product of spacing function and the likelihood function. Additionally, the highest posterior density (HPD) credible intervals are compared with the approximate confidence intervals (CIs) for the parameters of the model that were derived using both traditional approaches. Moreover, percentile bootstrap technique is utilized to compute confidence intervals. Numerical comparisons are presented of the proposed estimators with respect to various criteria quantities using Monte Carlo simulations. Further, using different optimality criteria, an optimal censoring scheme has been suggested. Besides, one-sample and two-sample prediction problems based on observed sample and predictive intervals under Bayesian framework are discussed. Finally, to demonstrate the proposed methodology in a real-life scenario, maximum flood level data is considered to show the applicability of the proposed methods.

Prophylaxis to Prevent Cardiotoxicity in Patients Receiving Anthracycline for Breast Cancer: A Combined Bayesian and Frequentist Network Meta-Analysis of Randomised Controlled Trials

The benefits in survivorship gained with anthracycline (ANT)-based chemotherapies for breast cancer are unfortunately mitigated for some patients by irreversible cardiotoxicity. Randomised controlled trials (RCTs) have explored multiple cardioprotection options, however, it remains unclear which drug is most effective in preserving left ventricular ejection fraction (LVEF). This study aimed to perform a systematic review and network meta-analysis, using Bayesian and frequentist approaches, of RCTs evaluating cardioprotective agents. Two authors searched four databases (CENTRAL, Cochrane Reviews, MEDLINE, SCOPUS), to find RCTs evaluating cardioprotective agents. Trial populations were limited to patients with breast cancer without prior ANT exposure. The primary outcome was mean LVEF change pre and post ANT dosing. Our primary analysis utilised a Bayesian approach, while our sensitivity analysis used frequentist methodology (Prospero registration number CRD42020199580). From 4,007 search results, we identified 12 RCTs, with their various trial arms considered separately-nine beta-blocker (BB), two angiotensin-converting enzyme inhibitor /angiotensin receptor blockers [(AA)+BB=AABB], one AA, one spironolactone, one statin-evaluating 1,126 patients (age 50.5 years). Bayesian network meta-analysis showed no difference in LVEF preservation between AA (1.3%, 95% credible interval [-0.20, 2.9]), BB (0.77, [-0.21, 1.8]), AABB (0.84 [-1.1, 2.8]), spironolactone (0.72, [-2.3, 3.7]) or statin (0.60, [-2.4, 3.6]) when compared against placebo. However, the frequentist analysis showed benefits from using AA (mean difference, 1.32% [0.32, 2.33]) and BB (mean difference, 0.76% [0.12, 1.4]). There is insufficient evidence to support prophylactic cardioprotection to prevent EF reduction. However, frequentist analysis suggested that AA or BBs provide cardioprotection. Thus, for those already on other anti-hypertensives, switching to AA or BBs could be considered.

AovBay: An R Package for Application and Visualization of Parametric Non-parametric and Bayesian ANOVA

The analysis of variance is a statistical technique widely used in the design of experiments and different research areas. It has been modeled using a classical or frequentist approach. With the computational power that is currently available, the Bayesian approach is an essential statistical tool related to hypothesis testing. However, conformity with classical techniques, ignorance of Bayesian statistics, and lack of easy-to-use software are obstacles to its frequent application. In this work, the use of a reproducible statistical package in R is proposed. It presents options to perform an analysis of variance in a classical (frequentist) and Bayesian way when the assumptions of the frequentist approach are not met or when a level of more specific inference such as quantifying the evidence provided by a data set for a given hypothesis, with the possibility of contributing to the understanding of the rejection or not of the statistical hypotheses raised, through interactive graphics presented in an emerging Shiny panel.

BAYESIAN APPROACH IN NEURAL NETWORKS

With the development of Deep learning, neural networks have become very popular. But nowadays neural networks which are being used are based on standard approach of Statistics. The main purpose of this work is to present Bayesian approach, highlight the main differences between Bayesian and frequentist approaches, their principles The problem was set to show in which cases Bayesian neural networks can be more preferable. The purpose was achieved through the following stages. The main steps of neural networks are presented. After it the fundaments of Bayesian approach are described. As Bayesian approach is a common concept, not just an inference used in neural networks, initially author speaks about this approach generally and then tells about its usage in neural networks. After that the main advantages and limitations of Bayesian networks in Deep learning are spoken about: which problems they can solve. The main conclusion of this paper is that Bayesian approach could be used to avoid overfitting. However, it is essential to understand the specific conditions under which Bayesian neural networks are preferable.

Frequentist and Bayesian approach for the generalized logistic lifetime model with applications to air-conditioning system failure times under joint progressive censoring data

Frequentist and Bayesian approach for the generalized logistic lifetime model with applications to air-conditioning system failure times under joint progressive censoring data

Implementing measurement error models with mechanistic mathematical models in a likelihood-based framework for estimation, identifiability analysis and prediction in the life sciences.

Throughout the life sciences, we routinely seek to interpret measurements and observations using parametrized mechanistic mathematical models. A fundamental and often overlooked choice in this approach involves relating the solution of a mathematical model with noisy and incomplete measurement data. This is often achieved by assuming that the data are noisy measurements of the solution of a deterministic mathematical model, and that measurement errors are additive and normally distributed. While this assumption of additive Gaussian noise is extremely common and simple to implement and interpret, it is often unjustified and can lead to poor parameter estimates and non-physical predictions. One way to overcome this challenge is to implement a different measurement error model. In this review, we demonstrate how to implement a range of measurement error models in a likelihood-based framework for estimation, identifiability analysis and prediction, called profile-wise analysis. This frequentist approach to uncertainty quantification for mechanistic models leverages the profile likelihood for targeting parameters and understanding their influence on predictions. Case studies, motivated by simple caricature models routinely used in systems biology and mathematical biology literature, illustrate how the same ideas apply to different types of mathematical models. Open-source Julia code to reproduce results is available on GitHub.

Effects of Anesthetics on Cardiac Repolarization in Adults: A Network Meta-Analysis of Randomized Clinical Trials.

Prolongation of cardiac repolarization, especially the heart rate-corrected QT (QTc) interval, is associated with life-threatening dysrhythmias. This study aimed to identify the anesthetic with the lowest risk of prolonging cardiac repolarization and provide guidance for anesthesia management in patients with cardiac diseases or long QT syndrome. Randomized controlled trials (RCTs) comparing the effects of anesthetics on cardiac repolarization indices were searched for in multiple databases. The primary outcome was QTc; and the secondary outcomes were other repolarization indices. A network meta-analysis was conducted using a frequentist approach and registered with the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD42022304970). Thirteen RCTs investigating 953 adults with normal QTc interval and without cardiovascular diseases were included. Direct meta-analyses found that propofol had less influence than sevoflurane (95% confidence interval (CI): 16.10, 33.54) and desflurane (95% CI: 4.85, 35.36), and sevoflurane had less influence than desflurane (95% CI: 6.96, 19.39) on QTc prolongation. Network analysis found that propofol had less influence than sevoflurane (95% CI: 17.78, 29.63), halothane (95% CI: 11.29, 41.24), desflurane (95% CI: 23.79, 39.88), and isoflurane (95% CI: 20.11, 46.10), and sevoflurane had less influence than desflurane (95% CI: 0.43, 15.82) on QTc prolongation. The rank order of cumulative ranking curve analysis was propofol (100%), sevoflurane (63.8%), halothane (49.5%), desflurane (21.1%), and isoflurane (15.6%). The direct meta-analysis found that propofol had less influence than sevoflurane on QT prolongation (95% CI: 23.12, 57.86). Other secondary outcomes showed no conclusive findings. This meta-analysis found that propofol had a minimal effect on QTc prolongation, followed by sevoflurane and desflurane in adults with normal QTc interval and without cardiovascular diseases. Propofol is the best anesthetic for adult patients with long QT syndrome or cardiac diseases, but still needs more robust evidence.

Reliability of ultrasonic tomography in detecting grouting defects in post-tensioned structures by PoD curves

The present research aims to assess the reliability of Ultrasonic Tomography (UT) of void detection in post-tensioning ducts due to inappropriate execution of grouting. The methodology is based on the construction of the Probability of Detection (PoD) curves for this specific non-destructive technique, by relating the probability of success to identify a specified ungrouted region. The PoD curves are built on data collected during a testing campaign in the laboratory, under fully known grouting conditions. Six walls containing prestressing strands and bars in metallic and plastic ducts with simulated voids of different sizes were designed and executed. Both water-filled (infiltration) and empty voids (ungrouted region) were simulated. The classical frequentist and Bayesian approaches are used to derive the PoD curves. Since the available experimental data are limited, the Bayesian approach provides more reliable data and is, thus, proposed as the best alternative. It is finally concluded that the minimum size of the voids in metallic ducts that UT can detect with 95% confidence and 90% probability, is around 30 cm and this value will be used in future inspection campaigns in posttensioned bridges.

Different Outcomes According to Needling Point Location Used in Sham Acupuncture for Cancer-Related Pain: A Systematic Review and Network Meta-Analysis.

Numerous acupuncture studies have been conducted on cancer-related pain; however, its efficacy compared to sham acupuncture remains controversial. We confirmed whether the outcome of acupuncture differs according to the needling points of sham acupuncture for cancer-related pain. We searched 10 databases on 23 May 2023 to screen acupuncture trials using sham acupuncture or waiting list as controls for cancer-related pain. Sham acupuncture was classified into two types, depending on whether the needling was applied at the same locations as verum acupuncture (SATV) or not (SATS). A network meta-analysis (NMA) was performed on the basis of a frequentist approach to assess pain severity. Eight studies (n = 574 participants) were included in the review, seven of which (n = 527 participants) were included in the NMA. The pain severity was not significantly different between SATV and verum acupuncture, but verum acupuncture significantly improved pain severity compared to SATS. The risk of bias affecting the comparisons between the verum and sham acupuncture was generally low. Previous acupuncture trials for cancer-related pain showed differing outcomes of sham and verum acupuncture, depending on the needling points of sham acupuncture. The application of SATV cannot be considered a true placebo, which leads to an underestimation of the efficacy of verum acupuncture.

Use of recombinant malate dehydrogenase (MDH) and superoxide dismutase (SOD) [Cu[sbnd]Zn] as antigens in indirect ELISA for diagnosis of bovine brucellosis

The objective of this study was to validate an indirect enzyme-linked immunoassay (iELISA) using the recombinant proteins, malate dehydrogenase (MDH) and superoxide dismutase (SOD) [CuZn], as antigens and to evaluate its ability to discriminate antibodies produced by vaccination from those induced by infection in the diagnosis of bovine brucellosis. Sera from six groups were evaluated: G1 – culture-positive animals (52 serum samples) (naturally infected); G2 - non-vaccinated animals (28 serum samples) positive in RBT (Rose Bengal test) and 2ME (2-mercaptoethanol test) selected from brucellosis-positive herds; G3 - animals from a brucellosis-free area (32 serum samples); G4 - S19 vaccinated heifers (114 serum samples); G5 – RB51 vaccinated heifers (60 serum samples); G6 - animals inoculated with inactivated Yersinia enterocolitica O:9 (42 serum samples). Diagnostic sensitivity (DSe) and diagnostic specificity (DSp) were estimated using the frequentist approach and the confidence interval (CI) (95%) calculated by the Clopper-Pearson (exact) method. The DSe for iELISA_MDH in the G1 group was 71.7% (CI 95%: 57.6–83.2%) and for the G2 100.0% (CI 95%: 87.7–100.0%), whereas the DSp was 84.4% in the G3 (CI 95%: 67.2–94.7%). For the iELISA_SOD the DSe was estimated 67.3% for the G1 (CI 95%: 52.9–79.7%) and 71.4% for G2 (CI 95%: 51.3–86.8%), while the DSp for G3 was 87.5% (CI 95%: 71.0–96.5%). iELISA_MDH and iELISA_SOD showed potential to be used in the diagnosis of infected animals, increasing the range of serological tests available for the diagnosis of bovine brucellosis, with the advantage of being S-LPS-free. However, none of the tests could differentiate between infection and vaccination.

Frequentist model averaging for analysis of dose-response in epidemiologic studies with complex exposure uncertainty.

In epidemiologic studies, association estimates of an exposure with disease outcomes are often biased when the uncertainties of exposure are ignored. Consequently, corresponding confidence intervals (CIs) will not have correct coverage. This issue is particularly problematic when exposures must be reconstructed from physical measurements, for example, for environmental or occupational radiation doses that were received by a study population for which radiation doses cannot be measured directly. To incorporate complex uncertainties in reconstructed exposures, the two-dimensional Monte Carlo (2DMC) dose estimation method has been proposed and used in various dose reconstruction efforts. The 2DMC method generates multiple exposure realizations from dosimetry models that incorporate various sources of errors to reflect the uncertainty of the dose distribution as well as the uncertainties in individual doses in the exposed population. Traditional measurement-error model approaches, typically based on using mean doses in the dose-exposure analysis, do not fully account exposure uncertainties. A recently developed statistical approach that overcomes many of these limitations by analyzing multiple exposure realizations in relation to disease risk is Bayesian model averaging (BMA). The analytic advantage of the BMA is its ability to better accommodate complex exposure uncertainty in the risk estimation, but a practical. Drawback is its significant computational complexity. In this present paper, we propose a novel frequentist model averaging (FMA) approach which has all the analytical advantages of the BMA method but is much simpler to implement and computationally faster. We show in simulations that, like BMA, FMA yields 95% confidence intervals for association parameters that close to 95% coverage rate. In simulations, the FMA has shorter length of CIs than those of another frequentist approach, the corrected information matrix (CIM) method. We illustrate the similarities in performance of BMA and FMA from a study of exposures from radioactive fallout in Kazakhstan.

Dynamic and static predictive modelling of psychotherapy outcome-Comparison of two statistical approaches.

Improving prediction abilities in the therapy process can increase therapeutic success for a variety of reasons, such as more personalised treatment or resource optimisation. The increasingly applied methods of dynamic prediction seem to be very promising for this purpose. Prediction models are usually based on static approaches of frequentist statistics. However, the application of this statistical approach has been widely criticised in this research area. Bayesian statistics has been proposed in the literature as an alternative, especially for the task of dynamic modelling. In this study, we compare the performance of predicting therapy outcome over the course of therapy between both statistical approaches. Based on a sample of 341 patients, a logistic regression analysis was performed using both statistical approaches. Therapy success was conceptualised as reliable pre-post improvement in brief symptom inventory (BSI) scores. As predictors, we used the subscales of the Outcome Questionnaire (OQ-30) and the Helping Alliance Questionnaire (HAQ) measured every fifth session, as well as baseline BSI scores. The influence of the predictors during therapy differs between the frequentist and the Bayesian approach. In contrast, predictive validity is comparable with a mean area under the curve (AUC) of 0.76 in both model types. Bayesian statistic provides an innovative and useful alternative to the frequentist approach in predicting therapy outcome. The theoretical foundation is particularly well suited for dynamic prediction. Nevertheless, no differences in predictive validity were found in this study. More complex methodology as well as further research seems necessary to exploit the potential of Bayesian statistics in this area.

Bayesian survival extrapolation for cost-effectiveness analysis: a case study of RELAY for ramucirumab in combination with erlotinib in the treatment of non-small-cell lung cancer

Aim Increasing trend for progression-free survival (PFS)-based primary endpoint in oncology has led to lack of mature overall survival (OS) data at the time of approval. To address this evidence gap in economic evaluations, we used a joint Bayesian approach to predict survival outcomes using immature OS data from the RELAY trial. Methods Patient data from RELAY and systematic literature review (SLR) of phase 3 randomized clinical trials with hazard ratio (HR) estimates of mature PFS and immature OS were considered. OS and PFS were analyzed individually using a univariate model; bivariate analysis was performed using a joint model based on modified Bayesian normal induced copula estimation model. First, a Bayesian univariate model incorporated informative priors based on predicted HR and acceleration factor for OS and PFS. Second, a Bayesian-based joint model of RELAY PFS and OS data was based on the correlation between PFS and OS established in trials of similar populations. Marginal distribution of PFS was used to estimate the same for OS. Results Publications (N = 122) of first-line treatments in patients with epidermal growth factor receptor (EGFR)-mutated non-small cell lung cancer were identified in the SLR, of which 36 trials were linked to RELAY. Twenty-six trials with HR data were used. The univariate model could predict OS with reduced uncertainty compared with the frequentist approach. In the joint model, the marginal OS distribution borrowed strength from the marginal PFS distribution through the established correlation coefficient. Limitations Bayesian approach was successfully used in RELAY analysis but may not be universally applied to oncology trials due to the different associations of OS and PFS and different trial patient populations. Conclusions We demonstrated that both the univariate and joint Bayesian models reduced uncertainty in predicting OS compared to frequentist method. The methodology introduced here will have potential applications in clinical decision-making for other oncology trials.

Comparative efficacy and tolerability of probiotic, prebiotic, and synbiotic formulations for adult patients with mild-moderate ulcerative colitis in an adjunctive therapy: A network meta-analysis

Background & aimsProbiotics, prebiotics, and synbiotics (PPS) have been widely used as adjuvant treatments in patients with ulcerative colitis (UC) in recent years. However, the most effective formulations of PPS have yet to be identified. We thus aimed to compare the efficacy and tolerability of different PPS formulations for mild-moderate UC. MethodsWe searched PubMed, Embase, Web of Science, and Cochrane CENTRAL from inception to June 24, 2023 for double-blind randomized controlled trials. We used a frequentist approach in random-effects models for network meta-analysis and the Grading of Recommendations Assessment, Development, and Evaluation approach to evaluate the certainty of evidence. ResultsWe analysed data from 20 trials involving 1153 patients. The combinations of specific strains of Lactobacillus and Bifidobacterium (CLB) (odds ratio (OR), 3.85; 95% confidence interval (CI), 1.40-10.60; low certainty) and combinations of specific strains of Lactobacillus, Bifidobacterium, and Streptococcus (CLBS) (OR, 2.20; 95% CI, 1.47-3.28; low certainty) significantly increased the clinical remission rate in intention-to-treat analysis (ITT) when compared to placebo. Similarly, compared with placebo, the two combinations significantly reduced clinical activity scores (standardized mean difference (SMD), -1.17 (95% CI, -1.68 to -0.65), low certainty; and SMD, -1.33 (95% CI, -1.81 to -0.86), low certainty, respectively). Hierarchical cluster analyses showed the two combinations formed clusters with high efficacy (clinical remission in ITT and clinical activity score) and tolerability (withdrawal due to worsening symptoms) within 12 weeks. ConclusionIn this systematic review, we found CLB and CLBS demonstrated a clinical benefit in adjuvant treatments, with a comparable tolerability and safety profile to placebo. Further trials are needed.Trial registration number: CRD42022344905

On classical and Bayesian inference for bivariate Poisson conditionals distributions: theory, methods and applications

Bivariate count data arise in several different disciplines (epidemiology, marketing, sports statistics, etc., to name but a few) and the bivariate Poisson distribution which is a generalization of the Poisson distribution plays an important role in modeling such data. In this article, we consider the inferential aspect of a bivariate Poisson conditionals distribution for which both the conditionals are Poisson but the marginals are typically non-Poisson. It has Poisson marginals only in the case of independence. It appears that a simple iterative procedure under the maximum likelihood method performs quite well as compared with other numerical subroutines, as one would expect in such a case where the MLEs are not available in closed form. In the Bayesian paradigm, both conjugate priors and non-conjugate priors have been utilized and a comparison study has been made via a simulation study. For illustrative purposes, a real-life data set is re-analyzed to exhibit the utility of the proposed two methods of estimation, one under the frequentist approach and the other under the Bayesian paradigm.