Glad to be involved in more #clinicaltrials using our ordinal longitudinal ordinal Markov model to obtain a high-powered endpoint. See the paper's #Statistics section and hbiostat.org/endpoint for more. Ordinal outcome is assessed hourly in this case.

1/ That "immortal time" is so frequent in survival analyses for #causalinference is fascinating. Because "immortal time" doesn't exist in the data, *we* create it when misanalyzing data. Our paper pubmed.ncbi.nlm.nih.gov/39494894/ summarizes why immortal time arises & how to prevent it.

2 weeks away! CAUSALab Methods Series Karolinska Institutet resumes w/ Nan van Geloven (Universiteit Leiden) presentation on predictions under interventions. 📆 January 28, 2025 ⏰ 15.00 CEST/9.00 ET 📍 All Methods talks virtual Register now 👇 stats.sender.net/forms/e7JD1d/v… #causalinference

New paper ✨ Our systematic review of #machinelearning in the prediction of opioid-associated adverse events published in Nature Digital Medicine npj Digital Medicine nature.com/articles/s4174… Led by Carlos Ramírez, with @AuriolesJose David A Jenkins Centre for Epidemiology Versus Arthritis UoM Musculoskeletal

Tutorials in Biostatistics Estimation of Diagnostic Test Accuracy Without Gold Standards. Ao Sun, Xiao-Hua Zhou. Statistics in Medicine. onlinelibrary.wiley.com/doi/10.1002/si…

The R rms package, now 34 years old, has a massive update. A key new #Statistics capability is profile likelihood confidence intervals and LR tests 4 general contrasts. And for ordinal regression the number of model intercepts is now unlimited. hbiostat.org/r/rms #RStats

Propensity Score Matching: should we use it in designing observational studies? Fei Wan. BMC Medical Research Methodology. bmcmedresmethodol.biomedcentral.com/articles/10.11…

Foundation models in bioinformatics academic.oup.com/nsr/advance-ar… 🧬🖥️🧪

Data Scientist vs. AI Engineer (Generative AI Edition) I've been studying AI for 18 months. This is what I discovered about the rise of this new role:

Invited Commentary: Why use methods that require proportional hazards? Mats J Stensrud, Miguel A Hernàn. American Journal of Epidemiology. doi.org/10.1093/aje/kw…

🚨#Tofacitinib ⬆️ #cardiovascular & #cancer risk in RA. But how relevant is this risk to people with #AxialSpondyloarthritis or #PsoriaticArthritis? This study did not find⬆️risk of JAK compared to TNF/IL17 inhibitors in SpA. Funded by PAPAA sciencedirect.com/science/articl…

🚀 fastml 0.5.0 is now on CRAN! 🎉 Train & evaluate multiple ML models with ease! Supports xgboost, random forest, SVM, LightGBM, elastic net, naive Bayes, MLP, and feature importance, SHAP values & more! #RStats #MachineLearning #DataScience cran.r-project.org/web/packages/f…

Read the book for free here: christophm.github.io/interpretable-…

In survival analysis, Restricted Mean Survival Time (RMST) offers a robust alternative to hazard ratios, especially when the proportional hazards assumption doesn’t hold. #DataScience #Biostatistics #Research #Science ahajournals.org/doi/10.1161/CI…

The #RStats qreport package for quarto reports has a new function makegvflow that allows users to easily make graphviz flow diagrams by specifying plain text with indenting to define the hierarchy. An example is at hbiostat.org/rflow/rformat#…

Regression imputation is a powerful method for handling missing data by predicting missing values based on relationships with other variables. Two widely used approaches, deterministic regression imputation and stochastic regression imputation, differ in how they handle

Frank Harrell Another reason to embrace longitudinal ordinal models! onlinelibrary.wiley.com/doi/10.1002/si…

Principal Component Analysis (PCA) effectively reduces complexity in data sets by condensing dimensions and extracting key patterns. Here's a straightforward guide on implementing PCA using R: 1️⃣ Prepare Your Data: Begin by loading your data. Ensure all variables are scaled or

Evaluation of the comparative risk of severe constipation by opioid type and dose in patients with non-cancer pain admitted to hospital. Read the full study by Meghna Jani Will Dixon Belay Birlie Mehreen @Carlosraul_ram here: bmcmedicine.biomedcentral.com/articles/10.11…

New book on Bayesian Meta-Analysis with Stan code: statmodeling.stat.columbia.edu/2025/07/15/a-b…