Article
Mathematical & Computational Biology
Clara Dominguez Islas, Kenneth M. Rice
Summary: Bayesian methods are a natural choice for combining evidence in meta-analyses, but their sensitivity to the choice of prior distribution, especially for parameters describing heterogeneity, is less attractive. A recent non-Bayesian approach to fixed-effects meta-analysis provides new ways of estimating an average effect and the variability around it. In this paper, we describe the Bayesian equivalents of these results, demonstrating that Bayesian inference on fixed-effects parameters is more stable and less sensitive than standard random-effects approaches.
RESEARCH SYNTHESIS METHODS
(2022)
Article
Mathematical & Computational Biology
Masayuki Henmi, Satoshi Hattori, Tim Friede
Summary: This paper proposes an improved method for estimating the overall effect in meta-analysis, particularly when dealing with only a few studies, which outperforms other methods in terms of coverage probabilities.
RESEARCH SYNTHESIS METHODS
(2021)
Article
Health Care Sciences & Services
Moritz Felsch, Lars Beckmann, Ralf Bender, Oliver Kuss, Guido Skipka, Tim Mathes
Summary: Meta-analysis is a method used to summarize the results of multiple studies. Standard methods for meta-analysis have limitations when there are only a few studies available, so alternative methods are needed. The common-rho beta-binomial model has shown good results in situations with sparse data or few studies, but it ignores the pairing of treatment and control arms in each study. This study extended the model to respect randomization. The simulation study found that the common-rho beta-binomial model is a good option for meta-analysis of very few studies.
BMC MEDICAL RESEARCH METHODOLOGY
(2022)
Article
Health Care Sciences & Services
Chang Xu, Luis Furuya-Kanamori, Liliane Zorzela, Lifeng Lin, Sunita Vohra
Summary: Researchers often struggle with how to handle zero-events in evidence synthesis, as improper handling can lead to research waste and misguidance in healthcare practice. This study proposes a framework to guide researchers in better dealing with zero-events in meta-analysis, classifying different types of zero-events studies and providing applicable methods under each subtype.
JOURNAL OF CLINICAL EPIDEMIOLOGY
(2021)
Article
Health Care Sciences & Services
Mohammed Baragilly, Brian Harvey Willis
Summary: Tailored meta-analysis uses setting-specific knowledge to constrain the possible values for a test's sensitivity and specificity, providing more reliable estimates.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2022)
Review
Psychology, Multidisciplinary
Simone Battista, Arianna Lazzaretti, Ilaria Coppola, Luca Falsiroli Maistrello, Nadia Rania, Marco Testa
Summary: This study synthesizes qualitative research on the experience of people with migraine, revealing the negative impact of migraine on various aspects of life, including family, work, and social relationships. It also highlights the emotional distress experienced by migraine sufferers and the coping strategies they use. The findings emphasize the need to raise awareness of this disabling condition in society and improve the quality of care provided to migraine patients from a social and health-policy perspective.
FRONTIERS IN PSYCHOLOGY
(2023)
Article
Multidisciplinary Sciences
Rebeca Mora-Castro, Marcela Alfaro-Cordoba, Marcela Hernandez-Jimenez, Mauricio Fernandez Otarola, Michael Mendez-Rivera, Didier Ramirez-Morales, Carlos E. Rodriguez-Rodriguez, Andres Duran-Rodriguez, Paul E. Hanson
Summary: Experimental observations on different genera of parasitoid wasps suggest that the BOB color pattern may serve as a warning signal, attracting the attention of spider predators.
Article
Mathematics, Interdisciplinary Applications
Yongnam Kim, Peter M. Steiner
Summary: This paper translates the conventional algebraic formalization of fixed effects (FE) and random effects (RE) models into causal graphs, providing intuitive explanations on how they consider different data-generating models and eliminate unmeasured confounding bias. The visualized causal graphs show that FE and RE estimators offset confounding bias by creating new non-causal paths and associations of opposite sign, in contrast to standard regression or matching estimators. Despite their similar mechanisms, subtle differences highlighted in the graphs can result in different biases in observational studies.
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY
(2021)
Article
Mathematics, Applied
Nanami Taketomi, Takeshi Emura
Summary: This article fills the research gap by clarifying the conditions for consistency of the common mean estimator in fixed-effect meta-analyses. Five theorems are devised to state the regularity conditions for the estimator to be consistent. These theorems are novel applications of large sample theory to meta-analyses. Numerical illustrations and real datasets demonstrate the practical consequences of the theorems. The article concludes that the common mean estimator can be inconsistent under certain conditions in real meta-analyses.
Article
Mathematical & Computational Biology
Spencer Hansen, Kenneth Rice
Summary: The paper presents methods for exact statistical tests and confidence intervals for fixed-effects meta-analysis of proportions, which retain the interpretability of the underlying parameter of interest and can be implemented in straightforward software. The inference on the overall proportion is shown to be compatible with exact inference on heterogeneity of proportions.
RESEARCH SYNTHESIS METHODS
(2022)
Article
Health Care Sciences & Services
Humaira Hussein, Keith R. R. Abrams, Laura J. J. Gray, Sumayya Anwer, Sofia Dias, Sylwia Bujkiewicz
Summary: This study explores and extends NMA models that consider the differences in study designs and assesses their impact on effect estimates and uncertainty. The results show that hierarchical and bias adjustment models that account for study design are more appropriate in NMA, providing more accurate estimates and uncertainty. Inclusion of non-randomised data allows for more generalizable inferences and the models accounting for study design allow for more appropriate modeling of complex data, preventing overly optimistic conclusions.
BMC MEDICAL RESEARCH METHODOLOGY
(2023)
Article
Mathematical & Computational Biology
Spencer Hansen, Kenneth Rice
Summary: Meta-analysis of associations between rare outcomes and binary exposures is important in drug side-effects studies. However, meta-analysis of the resulting 2 x 2 contingency tables has practical difficulties as analysts have to choose between exact inference and accounting for heterogeneity. The controversial Avandia meta-analysis highlights these challenges, with conflicting results depending on the methods used. In this article, the authors aim to resolve these difficulties by providing an exact and valid method that accounts for heterogeneity.
STATISTICS IN MEDICINE
(2023)
Article
Green & Sustainable Science & Technology
Festus Fatai Adedoyin, Andrew Adewale Alola, Festus Victor Bekun
Summary: This study investigates the relationship between alternative and sustainable energy sources, trade, income, and emissions in 27 selected European Union economies. The long-term results reveal that sustainable energy sources have a negative impact on pollutant emissions, while trade and income have a positive impact on carbon emissions, with trade having an insignificant impact. The findings suggest the need for increased use of sustainable energy sources to mitigate carbon emissions in the EU.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2021)
Review
Behavioral Sciences
Chunliang Feng, Simon B. Eickhoff, Ting Li, Li Wang, Benjamin Becker, Julia A. Camilleri, Sebastien Hetu, Yi Luo
Summary: The study found that different brain regions involved in various social interactions mainly mapped onto the default mode network, salience network, subcortical network, and central executive network, respectively involved in social cognition, motivation, and cognitive control. This discovery provides a heuristic integrative framework for understanding human social life from the perspective of component processes and network integration.
NEUROSCIENCE AND BIOBEHAVIORAL REVIEWS
(2021)
Article
Environmental Sciences
Nurhanis Syazni Roslan, Muhamad Saiful Bahri Yusoff, Karen Morgan, Asrenee Ab Razak, Nor Izzah Ahmad Shauki
Summary: This study aims to develop a concept of resilience in the context of physicians' experience through a meta-synthesis of relevant qualitative studies. Six key themes of resilience were identified: tenacity, resources, reflective ability, coping skills, control, and growth. Resilient physicians are determined, have control in their professional lives, reflect on adversity, utilize adaptive coping strategies, and believe that adversity provides an opportunity for growth. They are supported by individual and organizational resources.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Economics
Stephen Palmer, Isabelle Borget, Tim Friede, Don Husereau, Jonathan Karnon, Ben Kearns, Emma Medin, Elisabeth F. P. Peterse, Sven L. Klijn, Elisabeth J. M. Verburg-Baltussen, Elisabeth Fenwick, John Borrill
Summary: This article introduces an algorithm to help analysts decide whether flexible survival models are needed and which models should be chosen for testing. The algorithm consists of 8 steps and 4 questions, including systematic identification of relevant external data, using clinical expert input, considering future and observed hazard functions, assessing long-term survivorship potential, and presenting results from all plausible models. This algorithm provides a systematic, evidence-based approach to justify the selection of survival extrapolation models for cancer immunotherapies.
Article
Mathematical & Computational Biology
Marius Placzek, Tim Friede
Summary: In the era of precision medicine, adaptive designs play a significant role in clinical trials by offering flexibility and efficiency. This study proposes a strategy that combines blinded sample size recalculation and adaptive enrichment, and investigates the design characteristics through simulations.
BIOMETRICAL JOURNAL
(2023)
Article
Cardiac & Cardiovascular Systems
Muhammad Shahzeb Khan, Stefan D. Anker, Tim Friede, Ewa A. Jankowska, Marco Metra, Ileana L. Pina, Andrew J. S. Coats, Giuseppe Rosano, Bernard Roubert, Udo-Michael Goehring, Fabio Dorigotti, Josep Comin-Colet, Dirk J. Van Veldhuisen, Gerasimos S. Filippatos, Piotr Ponikowski, Javed Butler
Summary: This study aims to calculate the minimal clinically important difference (MCID) for change in exercise capacity in the 6-minute walk test (6MWT) in iron-deficient populations with heart failure and reduced ejection fraction (HFrEF). Through secondary analysis of two trials, the MCIDs for improvement and deterioration in the 6MWT at 12 and 24 weeks were determined.
JOURNAL OF CARDIAC FAILURE
(2023)
Article
Geriatrics & Gerontology
Ana Mazur, Britta Tetzlaff, Tina Mallon, Berit Hesjedal-Streller, Vivien Weiss, Martin Scherer, Sascha Koepke, Katrin Balzer, Linda Steyer, Tim Friede, Sebastian Pfeiffer, Eva Hummers, Christiane Mueller
Summary: To improve collaboration between general practitioners and nurses in nursing homes, researchers developed an intervention package called interprof ACT. However, the implementation of this intervention did not have a significant impact on hospitalisation rates and other clinical parameters among nursing home residents.
Article
Mathematical & Computational Biology
Sarah Friedrich, Tim Friede
Summary: Method comparisons are important for applied researchers to make informed choices. This paper investigates different approaches to comparison and aims to develop new methods by combining the advantages of these approaches.
BIOMETRICAL JOURNAL
(2023)
Article
Mathematical & Computational Biology
Roland G. G. Gera, Tim Friede
Summary: The increasing interest in subpopulation analysis in personalized medicine and targeted therapies has led to the development of various new trial designs and analysis methods. This paper proposes a trial design applicable to any set of composite populations and considers normally distributed endpoints and random baseline covariates. The study uses p-values calculated on subset levels and the inverse normal combination function to test treatment effects for composite populations, while also accounting for multiple testing using the closed testing procedure. The paper also derives critical boundaries for intersection hypothesis tests and provides simulations demonstrating the absence of practical relevant inflation of the type I error rate. The target power after sample size recalculation is typically met or close to being met.
BIOMETRICAL JOURNAL
(2023)
Book Review
Mathematical & Computational Biology
Tim Friede
BIOMETRICAL JOURNAL
(2023)
Article
Mathematical & Computational Biology
Christian Roever, Sibylle Sturtz, Jona Lilienthal, Ralf Bender, Tim Friede
Summary: In Bayesian meta-analysis, the specification of prior probabilities for heterogeneity is important, especially when there are few included studies. Utilizing empirical data from past analyses to inform the prior distributions is not straightforward. This study extends the commonly used normal-normal hierarchical model to infer a heterogeneity prior and presents simple approaches to fitting distributions to observed heterogeneity data.
STATISTICS IN MEDICINE
(2023)
Article
Mathematical & Computational Biology
Silvia Calderazzo, Sergey Tarima, Carissa Reid, Nancy Flournoy, Tim Friede, Nancy Geller, James Rosenberger, Nigel Stallard, Moreno Ursino, Marc Vandemeulebroecke, Kelly Van Lancker, Sarah Zohar
Summary: Clinical trials disruption is a significant issue in ending interventional studies. The COVID-19 pandemic has not only led to a surge in clinical research, but also a substantial disruption in non-COVID-19 trials, with about 80% being stopped or interrupted. This paper presents methods to compensate for the loss of information caused by trial disruptions by incorporating additional information from auxiliary data sources. These methods include using auxiliary data from the trial itself and incorporating information from external data sources using frequentist and Bayesian approaches. The methods are demonstrated using artificial data based on a clinical trial affected by the COVID-19 pandemic.
STATISTICS IN BIOPHARMACEUTICAL RESEARCH
(2023)
Article
Geriatrics & Gerontology
Markus S. Anker, Alessia Lena, Eric J. Roeland, Jan Porthun, Sebastian Schmitz, Sara Hadzibegovic, Philipp Sikorski, Ursula Wilkenshoff, Ann-Kathrin Froehlich, Luisa Valentina Ramer, Matthias Rose, Jan Eucker, Tienush Rassaf, Matthias Totzeck, Lorenz H. Lehmann, Stephan von Haehling, Andrew J. S. Coats, Tim Friede, Javed Butler, Stefan D. Anker, Hanno Riess, Ulf Landmesser, Lars Bullinger, Ulrich Keller, Johann Ahn
Summary: The patient-reported ability to walk 4 m and wash oneself is an independent predictor of survival and associated with decreased functional status in pre-terminal cancer patients.
JOURNAL OF CACHEXIA SARCOPENIA AND MUSCLE
(2023)
Article
Mathematical & Computational Biology
Eric S. Knop, Markus Pauly, Tim Friede, Thilo Welz
Summary: Analysts often overlook interaction terms in meta-regression models, which may result in biased conclusions. We demonstrate this through a reanalysis of a meta-regression study on acute heart failure. With a total of 285 studies, we examine the 1-year mortality rate and its association with study-level factors such as recruitment year and average age. Our findings highlight the importance of including possible confounders and interaction terms in mixed-effects meta-regression models to avoid erroneous inference.
RESEARCH SYNTHESIS METHODS
(2023)
Article
Cardiac & Cardiovascular Systems
Ursula Rauch-Kroehnert, Marianna Puccini, Marius Placzek, Jan Beyer-Westendorf, Kai Jakobs, Julian Friebel, Selina Hein, Mirko Seidel, Burkert Pieske, Steffen Massberg, Martin Witzenrath, Andreas Zeiher, Tim Friede, Stefan D. Anker, Ulf Landmesser
Summary: This study aimed to investigate the efficacy of therapeutic anticoagulation and prophylactic anticoagulation in COVID-19 patients. The results showed that therapeutic anticoagulation did not improve clinical outcome surrogates in patients with moderate to severe COVID-19 compared to prophylactic anticoagulation. Whether initial therapeutic doses of rivaroxaban are superior to thromboprophylaxis in COVID-19 patients with a high risk defined by d-dimer >2 ULN needs further confirmation.
CLINICAL RESEARCH IN CARDIOLOGY
(2023)
Article
Pharmacology & Pharmacy
Olympia Papachristofi, Bjoern Bornkamp, Melanie Wright, Tim Friede
Summary: Adaptive seamless trial designs have gained popularity in drug development, but making efficient decisions on adaptations in interim analyzes remains a challenge. This manuscript uses a case study to showcase potential adaptive features and the role of simulations in assessing design operating characteristics and decision rules. It also highlights the elements needed to support successful decision making and the importance of preplanning operational aspects.
PHARMACEUTICAL STATISTICS
(2023)
Article
Medicine, General & Internal
Britta Tetzlaff, Martin Scherer, Katrin Balzer, Linda Steyer, Sascha Koepke, Tim Friede, Indre Maurer, Clarissa E. Weber, Hans-Helmut Koenig, A. Konnopka, Thomas Ruppel, Ana Mazur, Eva Hummers, Christiane A. Mueller
Summary: The study aims to develop a care concept for improving person-centred interprofessional collaboration for people receiving home care. The methods include literature review, qualitative inquiries, quantitative study, and structured workshop. The findings will be used to elaborate and implement an interprofessional care concept.
Article
Clinical Neurology
Nicole Krause, Carlotta Derad, Barbara von Glasenapp, Karin Riemann-Lorenz, Herbert Temmes, Markus van de Loo, Tim Friede, Thomas Asendorf, Christoph Heesen
Summary: The study investigated the health behaviors and disease characteristics of newly diagnosed MS patients in Germany and found associations between these aspects. The results showed that the majority of participants had unhealthy lifestyle habits, such as smoking, obesity, and lack of physical activity. Higher disease severity scores were associated with higher lesion burden and relapse rates, while smoking was associated with higher lesion burden and unhealthy dietary habits. Higher disease severity scores and BMI were associated with lower quality of life. Additionally, lower quality of life was associated with a desire to optimize stress management, physical activity, and sleep behavior.
MULTIPLE SCLEROSIS AND RELATED DISORDERS
(2023)