Article
Astronomy & Astrophysics
Frank F. Deppisch, Graham Van Goffrier
Summary: Bayesian parameter inference techniques in neutrinoless double beta decay searches require a choice of prior distribution, which can strongly impact statistical conclusions. Least-informative priors are discussed to maximize information gain from an experimental setup in the parametrization of the lightest neutrino mass and an effective Majorana phase parameter.
Article
Psychology, Multidisciplinary
Christoph Koenig
Summary: The study aims to provide a new method for weighting informative prior distributions in Bayesian multiple regression models by combining sources of heterogeneity and a similarity measure ω. Through a comprehensive simulation study, the performance and behavior of the similarity-weighted informative prior distribution are investigated and compared to existing methods. The results offer applied researchers a means to specify accurate informative prior distributions.
FRONTIERS IN PSYCHOLOGY
(2021)
Article
Health Care Sciences & Services
Francesco Ungolo, Edwin R. van den Heuvel
Summary: This study introduces a joint model for addressing the problem of informative censoring in survival studies, utilizing latent variables and a fully Bayesian approach. Results suggest that ignoring informative censoring may lead to serious biases.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2022)
Article
Mathematical & Computational Biology
Omidali Aghababaei Jazi, Eleanor Pullenayegum
Summary: This article addresses the issue of selecting relevant variables and estimating coefficients in semiparametric regression models for longitudinal data with informative observation times. The authors propose a variable selection procedure suitable for estimation methods based on pseudo-score functions. The asymptotic properties of penalized estimators are investigated, and simulation studies are conducted to illustrate the theoretical results. The procedure is also applied to a real dataset from a randomized clinical trial.
STATISTICS IN MEDICINE
(2022)
Article
Endocrinology & Metabolism
Steven Craig, Cynthia Stretch, Farshad Farshidfar, Dropen Sheka, Nikolay Alabi, Ashar Siddiqui, Karen Kopciuk, Young Joo Park, Moosa Khalil, Faisal Khan, Adrian Harvey, Oliver F. Bathe
Summary: The clinical management of papillary thyroid cancer depends on accurate prognosis estimation. However, the standard prognostic methods based on clinicopathologic features have been found to be inaccurate. In this study, a machine learning algorithm was used to analyze 502 cases, leading to the identification of three distinct molecular subtypes with different biological features and implications for treatment strategies.
FRONTIERS IN ENDOCRINOLOGY
(2023)
Article
Medical Informatics
Jillian L. McKee, Michael C. Kaufman, Alexander K. Gonzalez, Mark P. Fitzgerald, Shavonne L. Massey, France Fung, Sudha K. Kessler, Stephanie Witzman, Nicholas S. Abend, Ingo Helbig
Summary: Using standardized EEG reports, we developed seizure prediction models for high-risk neonates, which can help in directing resource-intensive continuous EEG monitoring. This study highlights the importance of accurate seizure prediction in allocating limited EEG resources.
LANCET DIGITAL HEALTH
(2023)
Article
Engineering, Industrial
Pablo-Ramses Alonso-Martin, Ignacio Montes, Enrique Miranda
Summary: Human Reliability Analysis aims to address the hazardous consequences caused by human factors. We propose an alternative based on distortion models to quantify the influence of human factors and provide robust estimations.
Article
Statistics & Probability
Negera Wakgari Deresa, Ingrid Van Keilegom
Summary: Most existing copula models assume known parameters, but in reality, prior knowledge on this dependence parameter is often unavailable. We propose a novel model where the copula parameter does not need to be known, using a parametric copula model for the relation between survival time and censoring time. The model is shown to be identified, and estimators for the nonparametric cumulative hazard and finite-dimensional parameters are proposed.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Article
Health Care Sciences & Services
Mark J. Rutherford, Therese M. -L. Andersson, Tor Age Myklebust, Bjorn Moller, Paul C. Lambert
Summary: Ensuring fair comparisons of cancer survival statistics across population groups requires careful consideration of differential competing mortality due to other causes and adjusting for imbalances over groups in other prognostic covariates. The methodological approach of using standardised and reference adjusted metrics offers an appealing approach for future cancer survival comparison studies and routinely published cancer statistics. The non-parametric estimation approach through the use of weighting allows the ability to estimate comparable survival estimates without the need for statistical modelling.
BMC MEDICAL RESEARCH METHODOLOGY
(2022)
Article
Biodiversity Conservation
A. Cisterne, J. Pay, R. Crates, D. Stojanovic
Summary: Population and life history data is often lacking for many species, making research and threat assessment challenging. This study used population models to assess the impact of uncertain threatening processes in data-deficient Tasmanian masked owls. The study found that reproductive output, adult mortality, and juvenile mortality had the greatest impact on population size and growth. The study also simulated the effects of rodenticide exposure on the population and found that the magnitude of impacts depended on the choice of uncertain parameter.
GLOBAL ECOLOGY AND CONSERVATION
(2023)
Review
Nutrition & Dietetics
Savita A. Sandhu, Chloe A. Angel, Katrina L. Campbell, Ingrid J. Hickman, Helen L. MacLaughlin
Summary: This review found that there is extensive heterogeneity in the reporting of outcomes in randomized controlled trials investigating nutrition interventions for the management of complex chronic disease, with only a small number of outcomes being reported with relative consistency. There is also a limited scope of outcomes reported, with a notable lack of patient-reported outcomes. Additionally, poor agreement was observed between the outcomes reported in the RCTs and those recommended by the Core Outcome Sets (COS).
Article
Biochemical Research Methods
Michael Komodromos, Eric O. Aboagye, Marina Evangelou, Sarah Filippi, Kolyan Ray
Summary: This paper proposes an interpretable and scalable Bayesian proportional hazards model, referred to as sparse variational Bayes, for analyzing high-dimensional sparse survival data. The proposed method overcomes the high computational cost of traditional methods and offers a mechanism for variable selection via posterior inclusion probabilities. Extensive simulations demonstrate the comparable or better performance of the proposed method compared to state-of-the-art Bayesian variable selection methods.
Review
Food Science & Technology
Mariana Santos, Ana Isabel Rito, Filipa Nunes Matias, Ricardo Assuncao, Isabel Castanheira, Isabel Loureiro
Summary: Nutrient profiling is an important tool for improving public health nutrition and reducing the prevalence of noncommunicable diseases. Studies show that sodium, saturated fatty acids, and total sugars are the most commonly restricted nutrients, while fiber is the most commonly used qualifying nutrient.
TRENDS IN FOOD SCIENCE & TECHNOLOGY
(2021)
Article
Genetics & Heredity
Jerry Hoogenboom, Titia Sijen, Kristiaan J. van der Gaag
Summary: The STRNaming algorithm generates human-readable names for sequenced STR alleles by using a reference sequence at each locus and automatically assigning a unique, sequence-descriptive name that also includes the capillary electrophoresis allele number. Settings for STRNaming were established based on international preferences in the forensic community, ensuring that small changes in the sequence correspond to small changes in the allele name, aiding in recognizing stutter products and other sequence variants.
FORENSIC SCIENCE INTERNATIONAL-GENETICS
(2021)
Article
Health Care Sciences & Services
Andras Lanczky, Balazs Gyorffy
Summary: The web-based tool introduced in this study enables univariate and multivariate Cox proportional hazards survival analysis using data from various genomic studies, filling a significant gap in the field where no commonly used spreadsheet software or web server is available for this purpose.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Oncology
Therese M. -L. Andersson, Tor Age Myklebust, Mark J. Rutherford, Bjorn Moller, Melina Arnold, Isabelle Soerjomataram, Freddie Bray, D. Maxwell Parkin, Paul C. Lambert
Summary: This study examines the comparison of population-based cancer survival rates internationally and aims to improve the harmonization of cancer registry datasets for future benchmarking exercises. It emphasizes the importance of understanding and addressing international variations in outcomes among cancer patients.
BRITISH JOURNAL OF CANCER
(2022)
Article
Oncology
Therese M. -L. Andersson, Mark J. Rutherford, Tor Age Myklebust, Bjorn Moller, Melina Arnold, Isabelle Soerjomataram, Freddie Bray, Hazem Abd Elkader, Gerda Engholm, Dyfed Huws, Alana Little, Lorraine Shack, Paul M. Walsh, Ryan R. Woods, D. Maxwell Parkin, Paul C. Lambert
Summary: The study found that, except in Ireland, there were few long-term survivors in the early years of the ICBP SURVMARK-2 project. This finding is important for accurately estimating cancer patient survival.
CANCER EPIDEMIOLOGY
(2022)
Article
Health Care Sciences & Services
Yuliya Leontyeva, Hannah Bower, Oskar Gauffin, Paul C. Lambert, Therese M-L Andersson
Summary: Ignoring uncertainty in general population mortality rates has negligible impact on the SEs of 5-year RS and LLE when based on the whole population, but can have a larger impact when based on a smaller population for expected mortality rates estimation, particularly for marginal values.
BMC MEDICAL RESEARCH METHODOLOGY
(2022)
Article
Oncology
Paramita Dasgupta, Therese M-L Andersson, Gail Garvey, Peter D. Baade
Summary: There are differences in remaining life expectancy (RLE) among Aboriginal and Torres Strait Islander and other Australian patients with cancer, partly due to disparities in cancer and noncancer mortality. If the cancer survival disparities were removed, Aboriginal and Torres Strait Islanders diagnosed with cancer could gain more population life years.
CANCER EPIDEMIOLOGY BIOMARKERS & PREVENTION
(2022)
Article
Statistics & Probability
Nurgul Batyrbekova, Hannah Bower, Paul W. Dickman, Robert Szulkin, Paul C. Lambert, Therese M. -L. Andersson
Summary: When analyzing cohort data, it is common to only consider one time-scale and include other time-scales as fixed covariates. This simulation study demonstrates that only modeling one time-scale can lead to biased estimates of survival proportions and hazard ratios, especially when there are non-proportional hazards on the second time-scale. Including non-proportional hazards and interactions with the time-scale is advisable when modeling only one time-scale.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2022)
Article
Oncology
Maximilian Zoltek, Therese M-L Andersson, Erland Axelsson, Christel Hedman, Catharina Ihre Lundgren
Summary: This study investigated the relationship between health anxiety and TSH-suppression treatment in patients with differentiated thyroid cancer (DTC). The results showed that health anxiety appears to be slightly higher among DTC patients compared to the general population, but it is not clearly connected to the TSH-suppression treatment.
Article
Mathematical & Computational Biology
Nikolaos Skourlis, Michael J. Crowther, Therese M-L Andersson, Paul C. Lambert
Summary: In competing risks settings, using time since diagnosis as the timescale reduces bias in the cumulative incidence functions for cancer mortality and other cause mortality. However, using attained age as the timescale when modeling other cause mortality allows for more natural modeling.
BIOMETRICAL JOURNAL
(2022)
Article
Hematology
Cecilia Radkiewicz, Johanna B. Bruchfeld, Caroline E. Weibull, Mathias L. Jeppesen, Henrik Frederiksen, Mats Lambe, Lasse Jakobsen, Tarec C. El-Galaly, Karin E. Smedby, Tove Wasterlid
Summary: It is well established that males have a higher risk and poorer survival of most cancers, and this pattern also applies to lymphomas. In this study, researchers used the Swedish Lymphoma Register to investigate sex differences in lymphoma subtype incidence and excess mortality. The findings showed that men had a significantly higher incidence and trend toward higher mortality in most lymphoma subtypes.
AMERICAN JOURNAL OF HEMATOLOGY
(2023)
Article
Health Care Sciences & Services
Nurgul Batyrbekova, Hannah Bower, Paul W. Dickman, Anna Ravn Landtblom, Malin Hultcrantz, Robert Szulkin, Paul C. Lambert, Therese M-L. Andersson
Summary: This study introduces a method to model multiple time-scales in survival analysis, avoiding data splitting, saving time, and reducing the likelihood of errors.
BMC MEDICAL RESEARCH METHODOLOGY
(2022)
Article
Dermatology
Isabelle Krakowski, Henrike Habel, Kari Nielsen, Christian Ingvar, Therese M. L. Andersson, Ada Girnita, Karin E. Smedby, Hanna Eriksson
Summary: This study investigated the association between metformin use and survival in diabetic patients with cutaneous melanoma. The results showed that metformin use was associated with improved overall survival regardless of timing and followed a dose-response pattern. However, no association was found between metformin use and melanoma-specific survival.
BRITISH JOURNAL OF DERMATOLOGY
(2023)
Article
Oncology
Ida Wiken, Therese M-L Andersson, Cecilia Radkiewicz
Summary: This study investigates the impact of holidays on cancer incidence and survival rates in Sweden. It finds that cancer cases decline during holidays and increase post-holidays for certain types of cancer. The study also reveals higher mortality rates for cancers diagnosed during holidays, emphasizing the need for consistent cancer care throughout the year.
Article
Health Care Sciences & Services
Nikolaos Skourlis, Michael J. Crowther, Therese M-L. Andersson, Donghao Lu, Mats Lambe, Paul C. Lambert
Summary: This study uses multi-state models to explore different research questions and modelling choices in the analysis of repeated prescriptions data. The results show that more complex multi-state structures can effectively capture the intermittent nature of medication use.
BMC MEDICAL RESEARCH METHODOLOGY
(2023)
Article
Oncology
Cecilia Radkiewicz, Therese M-L Andersson, Jesper Lagergren
Summary: This study calculated years of life lost (YLL) due to the top 10 solid tumors causing the most deaths, using data from the Swedish National Cancer and Cause of Death Registers. The results showed that lung and colorectal cancer remained the leading causes of YLL, while pancreatic and breast cancer moved up in rank. Prostate cancer, on the other hand, moved down in rank. Women consistently lost more life years due to lung and pancreatic cancer, and the decline in colorectal cancer mortality was only reflected as a YLL decline in women. YLL is a simple and intuitive measure that expands our understanding of the cancer burden on society.
JNCI CANCER SPECTRUM
(2023)
Article
Oncology
Tor Age Myklebust, Bjarte Aagnes, Yngvar Nilssen, Mark Rutherford, Paul C. Lambert, Therese M. L. Andersson, Anna L. V. Johansson, Paul W. Dickman, Bjorn Moller
Summary: Background: Routine reporting of cancer patient survival is vital for monitoring healthcare effectiveness and prognosis. Objective: This study aims to explore the feasibility of automating the production of these statistics. Results: We obtained reliable estimates of survival rates for most cancer types. Discussion: Implementing new survival measures in routine publications may be challenging, but we propose a method for automating the production of these statistics.
BRITISH JOURNAL OF CANCER
(2023)