Malgorzata Roos

Background: Unrealistic recruitment planning at the design stage is a well-known cause of study failure and wasted resources. Therefore, it is crucial for trial designers to conduct a realistic RECruitment ANalysis (RECAN), taking into account the uncertainty when predicting how long it will take to recruit a required number of participants or how many participants will be recruited within a given time period. However, trial designers currently lack both an accessible tool for RECAN predictions and a standardized RECAN reporting scheme. A reporting scheme was recently proposed by Heesen et al. (2026), but has not yet been widely implemented. In addition, Carter (2004) proposed a feasible distributional RECAN approach, but the link to the source code is obsolete, highlighting the need for reimplementation.Methods: Addressing both of these issues, we use open-source R and Quarto software to reimplement, extend, and disseminate Carter’s distributional RECAN approach in the form of interactive open-source templates that are based on the recently standardized RECAN reporting scheme.Results: We developed RECAN templates for study designers to address the different possible use scenarios involving confidence considerations, constant and piecewise-constant recruitment rates over time. These templates provide distributional predictions, estimates, plots, and interpretations.Conclusions: The RECAN templates are available online to help trial designers make realistic and reproducible RECAN predictions during the design stage. The standardized template output could become an integral part of submissions to funding organizations, institutional review boards, and ethics committees, facilitating transparent decision-making and allowing more studies to reach their recruitment goals.

The two‐trials rule in drug regulation requires statistically significant results from two pivotal trials to demonstrate efficacy. However, it is unclear how the effect estimates from both trials should be combined to quantify the drug effect. Fixed‐effect meta‐analysis is commonly used but may yield confidence intervals that exclude the value of no effect even when the two‐trials rule is not fulfilled. We systematically address this by recasting the two‐trials rule and meta‐analysis in a unified framework of combined p ‐value functions, where they are variants of Wilkinson's and Stouffer's combination methods, respectively. This allows us to obtain compatible combined p ‐values, effect estimates, and confidence intervals, which we derive in closed‐form. Additionally, we provide new results for Edgington's, Fisher's, Pearson's, and Tippett's p ‐value combination methods. When both trials have the same true effect, all methods can consistently estimate it, although some show bias. When true effects differ, the two‐trials rule and Pearson's method are conservative (converging to the less extreme effect), Fisher's and Tippett's methods are anti‐conservative (converging to the more extreme effect), and Edgington's method and meta‐analysis are balanced (converging to a weighted average). Notably, Edgington's confidence intervals always asymptotically include the individual trial effects, while meta‐analytic confidence intervals shrink to a point at the weighted average effect. We conclude that all of these methods may be appropriate depending on the estimand of interest. We implement combined p ‐value function inference for two trials in the R package twotrials , allowing researchers to easily perform compatible hypothesis testing and effect estimation.

Cross-disciplinary collaborative research projects are often essential in many fields for making scientific progress, but they are notoriously difficult to conceptualize and design. The recent increased focus on reproducible science makes the design of such research projects even more challenging. We demonstrate that simple rules provided by three editorial articles from the series "Ten Simple Rules" greatly impacted the process of writing of our research proposal, leading to a design that can serve researchers from a wide range of disciplines to conceptualize their own collaborative projects that promote reproducible science.

Background: One of the emerging themes in epidemiology is the use of interval estimates. Currently, three interval estimates for confidence (CI), prediction (PI), and tolerance (TI) are at a researcher's disposal and are accessible within the open access framework in R. These three types of statistical intervals serve different purposes. Confidence intervals are designed to describe a parameter with some uncertainty due to sampling errors. Prediction intervals aim to predict future observation(s), including some uncertainty present in the actual and future samples. Tolerance intervals are constructed to capture a specified proportion of a population with a defined confidence. It is well known that interval estimates support a greater knowledge gain than point estimates. Thus, a good understanding and the use of CI, PI, and TI underlie good statistical practice. While CIs are taught in introductory statistical classes, PIs and TIs are less familiar. Results: In this paper, we provide a concise tutorial on two-sided CI, PI and TI for binary variables. This hands-on tutorial is based on our teaching materials. It contains an overview of the meaning and applicability from both a classical and a Bayesian perspective. Based on a worked-out example from veterinary medicine, we provide guidance and code that can be directly applied in R. Conclusions: This tutorial can be used by others for teaching, either in a class or for self-instruction of students and senior researchers.

In recent years, Bayesian meta‐analysis expressed by a normal–normal hierarchical model (NNHM) has been widely used for combining evidence from multiple studies. Data provided for the NNHM are frequently based on a small number of studies and on uncertain within‐study standard deviation values. Despite the widespread use of Bayesian NNHM, it has always been unclear to what extent the posterior inference is impacted by the heterogeneity prior (sensitivity ) and by the uncertainty in the within‐study standard deviation values (identification ). Thus, to answer this question, we developed a unified method to simultaneously quantify both sensitivity and identification (‐) for all model parameters in a Bayesian NNHM, based on derivatives of the Bhattacharyya coefficient with respect to relative latent model complexity (RLMC) perturbations. Three case studies exemplify the applicability of the method proposed: historical data for a conventional therapy, data from which one large study is first included and then excluded, and two subgroup meta‐analyses specified by their randomization status. We analyzed six scenarios, crossing three RLMC targets with two heterogeneity priors (half‐normal, half‐Cauchy). The results show that ‐ explicitly reveals which parameters are affected by the heterogeneity prior and by the uncertainty in the within‐study standard deviation values. In addition, we compare the impact of both heterogeneity priors and quantify how ‐ values are affected by omitting one large study and by the randomization status. Finally, the range of applicability of ‐ is extended to Bayesian NtHM. A dedicated R package facilitates automatic ‐ quantification in applied Bayesian meta‐analyses.

Meta‐analysis provides important insights for evidence‐based medicine by synthesizing evidence from multiple studies which address the same research question. Within the Bayesian framework, meta‐analysis is frequently expressed by a Bayesian normal‐normal hierarchical model (NNHM). Recently, several publications have discussed the choice of the prior distribution for the between‐study heterogeneity in the Bayesian NNHM and used several “vague” priors. However, no approach exists to quantify the informativeness of such priors, and thus, we develop a principled reference analysis framework for the Bayesian NNHM acting at the posterior level. The posterior reference analysis (post‐RA) is based on two posterior benchmarks: one induced by the improper reference prior, which is minimally informative for the data, and the other induced by a highly anticonservative proper prior. This approach applies the Hellinger distance to quantify the informativeness of a heterogeneity prior of interest by comparing the corresponding marginal posteriors with both posterior benchmarks. The post‐RA is implemented in the freely accessible R package ra4bayesmeta and is applied to two medical case studies. Our findings show that anticonservative heterogeneity priors produce platykurtic posteriors compared with the reference posterior, and they produce shorter 95% credible intervals (CrI) and optimistic inference compared with the reference prior. Conservative heterogeneity priors produce leptokurtic posteriors, longer 95% CrI and cautious inference. The novel post‐RA framework could support numerous Bayesian meta‐analyses in many research fields, as it determines how informative a heterogeneity prior is for the actual data as compared with the minimally informative reference prior.

To investigate the impact of pretreatment and conditioning on shear bond strength (SBS), surface free energy (SFE) and surface roughness (SR) between polyetheretherketone (PEEK) and cold-cured polymethylmethacrylate (PMMA). PEEK substrates (Dentokeep PEEK Disc, nt-trading) were air abraded with Al2O3 particles of different grain sizes applied with varying pressure at 1) 0.2 MPa - 50 µm Al2O3; 2) 0.4 MPa - 50 µm Al2O3; 3) 0.2 MPa - 110 µm Al2O3; 4) 0.4 MPa - 110 µm Al2O3; or 5) without air abrasion (n = 172/group). Surface properties were quantified using SFE and SR (n = 10/group), and scanning electron microscope imaging (n = 2/group). Substrates were conditioned with a) Visio.link (VL, Bredent); b) Scotchbond Universal (SU, 3M Oral Care); c) Bonding Fluid (BF, Schütz Dental); or d) without conditioning (WC; n = 40/subgroup) and bonded to the polymer (Futura Jet, Schütz Dental). SBS and fracture types were determined before and after 10,000 thermal cycles (n = 20/subgroup). Univariate ANOVA, Kruskal-Wallis test, Mann-Whitney U-test, Kaplan-Meier survival estimates, and Weibull distribution were computed (p < 0.05). Ciba-Geigy tables and the chi-squared test were used to analyze fracture type distributions. An increase in particle size and pressure resulted in similar or increased SBS, Weibull characteristic strength, and Weibull moduli (p < 0.001 - 0.046). The lowest results were observed for the control group (without air abrasion), while pretreatment with 0.4 MPa - 110 µm Al2O3 presented the highest values (p < 0.001). In comparison with the other conditioning procedures, VL showed high (p < 0.001 - 0.03), and SU and WC low SBS (p < 0.001 - 0.006). Although it did not influence SFE, an increase in particle size and pressure led to an increased SR (p < 0.001). Pretreatment with 0.4 MPa - 110 µm Al2O3 can be recommended to increase bonding properties between PEEK and PMMA. Application of adhesives such as VL can enhance SBS further.

Background: Kidney allograft survival continuously improved with introduction of novel immunosuppressants. However, also immunologically challenging transplants (blood group incompatibility and sensitized recipients) increase. Between 2006 and 2008, a new tailored immunosuppression scheme for kidney transplantation was implemented at the University Hospital in Zurich, together with an ABO-incompatible transplant program and systematic pre- and posttransplant anti-human leukocyte antigen (HLA) antibody screening by Luminex technology. This study retrospectively evaluated the results of this tailored immunosuppression approach with a particular focus on immunologically higher risk transplants. Methods: A total of 204 consecutive kidney transplantations were analyzed, of whom 14 were ABO-incompatible and 35 recipients were donor-specific anti-HLA antibodies (DSA) positive, but complement-dependent cytotoxicity crossmatch (CDC-XM) negative. We analyzed patient and graft survival, acute rejection rates and infectious complications in ABO-compatible versus -incompatible and in DSA positive versus negative patients and compared those with a historical control group. Results: Overall patient, death-censored allograft survival and non-death-censored allograft survival at 4 years were 92, 91 and 87%, respectively. We found that (1) there were no differences between ABO-compatible and -incompatible and between DSA positive and DSA negative patients concerning acute rejection rate and graft survival; (2) compared with the historical control group there was a significant decrease of acute rejection rates in sensitized patients who received an induction with thymoglobulin; (3) there was no increased rate of infection among the patients who received induction with thymoglobulin compared to no induction therapy. Conclusions: We observed excellent overall mid-term patient and graft survival rates with our tailored immunosuppression approach. Induction with thymoglobulin was efficient and safe in keeping rejection rates low in DSA positive patients with a negative CDC-XM.

Introduction: Continuous positive airway pressure (CPAP) is currently the treatment of choice for sleepiness in patients with obstructive sleep apnoea (OSA); however, adherence is often thought to be suboptimal. We investigated the effects of suboptimal CPAP usage on objective and subjective sleepiness parameters in patients with OSA.Material and methods: In this 2-week, parallel, double-blind, randomised controlled trial we enrolled moderate-to-severe OSA patients with excessive pre-treatment daytime sleepiness (Epworth sleepiness scale (ESS) score >10 points) who had suboptimal CPAP adherence over ≥12 months (mean nightly usage time 3–4 h). Patients were allocated through minimisation to either subtherapeutic CPAP (“sham CPAP”) or continuation of CPAP (“therapeutic CPAP”). A Bayesian analysis with historical priors calculated the posterior probability of superiority.Results: Between May, 2016 and November, 2018, 57 patients (aged 60±8 years, 79% male, 93% Caucasian) were allocated in total, and 52 who completed the study (50% in each arm) were included in the final analysis. The unadjusted ESS score increase was 2.4 points (95% CI 0.6–4.2, p=0.01) in the sham-CPAP group when compared to continuing therapeutic CPAP. The probability of superiority of therapeutic CPAP over sham CPAP was 90.4% for ESS, 90.1% for systolic blood pressure and 80.3% for diastolic blood pressure.Conclusions: Patients with moderate-to-severe OSA and daytime sleepiness are still getting a substantial benefit from suboptimal CPAP adherence, albeit not as much as they might get if they adhered more. Whether a similar statement can be made for even lower adherence levels remains to be established in future trials.