Darren Dahly PhD

Heredity is one of those words that still makes me feel stupid. I am unsure why, but I put it in the box with covariate factors, independent/dependent variables, and many other terms that someone needed a word for. Anyhow... Thanks for the explanation, Phil Kay! Another note on classifying factors as active or inactive is that there might be no detectable effect at the range you are testing. But what if you pushed the range to more extreme values? Make sure to follow: #DOEbyPhilKay for great explanations on #DOE

11

4 Comments

Check our data analysis series blog: A Comprehensive Guide to Correlation Network Graphs It covers: - The power of correlation network graphs in revealing hidden connections within metabolomic datasets. - Practical steps on creating and interpreting these network graphs for impactful research. - How these graphs can enhance collaboration with your research colleagues. Head over to our blog and unlock the power of network analysis: https://lnkd.in/gBqqKJCe #metabolomics #research #datascience #bioinformatics #networkanalysis

3

Insightful breakdown on one-tailed vs. two-tailed tests in biostatistics! Understanding these nuances is crucial for precision in statistical analysis. #Statistics #Biostatistics #Data #Statisticaltests

8

1 Comment

Maximizing data completeness and accuracy in #RWE research is essential to ensure that research findings are valid, reliable, and reproducible. With fit-for-purpose data, stakeholders can draw greater insights, advance medical innovation, enhance #regulatory processes, and make better informed decisions to improve #patientoutcomes. At ISPOR—The Professional Society for Health Economics and Outcomes Research 2024, Danielle Gentile, PhD and Richard Scott Swain, PhD, MPH, of Cardinal Health explored innovative strategies towards maximizing data completeness and accuracy, sharing their innovative approaches towards #RWD and physician-led chart collection in #oncology. Here we take a Deep Dive into the session “Innovative Strategies for Fit-for-Purpose RWE Research: Maximizing Data Completeness and Accuracy“, summarizing the key takeaways and discussions from the presentation. Read at The Evidence Base® #heor #healtheconomics #outcomesresearch #realworldevidence #realworlddata #regulation #regulatoryaffairs #healthdata #ISPORAnnual

7

A mistake to avoid in Clinical Research, Biostatistics and Data Science. Analysis without considering the subgroups can be a big mistake. Very often, it can lead to Simson's paradox (also called the Yule–Simpson effect) in which the overall data studied may show trends which actually diminish or even be completely reverse when subgroups are evaluated. What most don't really consider is that missing out on subgrouping may also lead to causal analysis failure as well and latent variable structure having more bias. This effect may be countered using different strategies from mainstream statistical ones to machine learning based ones of identifying and making the subgroups in the data : 1. Domain specific knowledge and subgroup labeling adding the biological context and identifying 2. Unsupervised Clustering as a type of unsupervised analysis used to identify data based subgroups 3. Any confounder correction principle will help, of course the randomization will help but other approaches may still be needed. 4. CART - classification and regression trees are one of my favorite aproaches in data driven subgroup identification. 5. Feature engineering based on both biological domains and data driven aproaches. 6. Last in this post but very important. AI aided subgorup analysis. AI models can be trained to identify the subgroups based on the biological labeling and the covariates of interest and using as a modeling augmentation aproach. You wont hear much about this one as its still a developing field. #datascience #biostatistics #clinicalresearch #analytics #ai #artificialintelligence #machinelearning #biology #ml #stats #biomedical

56

4 Comments

I'm doing a deep dive in #Causal #Inference literature - here are a few books I'm checking out to learn the fundamentals. 1️⃣ Judea Pearl, *Causality : Models, Reasoning and Inference*, 2000 A true masterpiece. A technical and deep exposition of Pearl's life work on Directed Acyclic Graphs (DAGs) as Structural Causal Models (SCMs) that got me started on my causal inference journey. His viewpoint is an alternative to the Neyman-Rubin causal model based on potential outcomes. 2️⃣ Guido W. Imbens and Donald B. Rubin, *Causal Inference for Statistics, Social, and Biomedical Sciences*, 2015   A true masterpiece. The most achieved and thorough exposition of the Neyman-Rubin causal model based on potential outcomes. It is an alternative to Pearl's DAG and SCM framework (see above). A beautiful book that I find myself going back to often, for its depth and breadth of insights into thinking about causal inference. 3️⃣ Judea Pearl, Madelyn Glymour & Nicholas P. Jewell, *Causal Inference in Statistics : A Primer*, 2016 A gentle introduction to Directed Acyclic Graphs (DAGs) and Structural Causal Models (SCMs) at about the undergraduate in statistics level. 4️⃣ Bill Shipley, *Cause and Correlation in Biology : A User's Guide to Path Analysis, Structural Equations and Causal Inference*, 2000 (Book) A well-written introduction to causal inference for biologists, with an emphasis of Structural Equation Models (SEMs) and Path Analysis. There is also a little bit of interesting history sprinkled in. I took a class with this professor (who just retired from a University close to my home town) in 2023! 5️⃣ Paul R. Rosenbaum, *Observational Studies*, 1995 (Book) A compact monograph by one of the inventors of the propensity score (PS) with a clear writing style. I've always enjoyed Rosenbaum's incisive perspectives on problems related to causality in observational studies! 6️⃣ Tyler J. VanderWeele, *Explanations in Causal Inference : Methods for Mediation and Interaction*, 2015   A thick, somewhat terse book by the leading researcher in mediation and interaction analysis in causal inference. This book is aimed at advanced users of causal inference methods, but it is truly unique in its depth of the subject! It is written by an epidemiologist but he also targets a social science audience. 🤔Do you have any other causal inference books or papers to recommend? Please, please, please, let me know! --- 👨‍💻The Github repo now contains RMarkdown notebooks with tutorials and blog posts, links to any resources available online, and is still growing! 📘My free biostatistics reading list can be found at https://lnkd.in/eCgMy_cZ. ⭐Star the repo if you want to receive any updates. ♻Please, share this post if you think it might be valuable to others 🙏 #statistics #biostatistics #data #science

79

23 Comments

Hey, statistical programmers in #clinicaltrials! Do you program clinical tables in R? Do you use the MS-Word (docx), RTF or HTML format? Can your favourite tool for farming tables do everything the 𝙛𝙡𝙚𝙭𝙩𝙖𝙗𝙡𝙚 + 𝙤𝙛𝙛𝙞𝙘𝙚𝙧 suite can? ⚡ numbers in table cells can be aligned with tabulation marks (no more space + fixed-width fonts!) to its decimal points! You can mix various types of tabs: decimal, left, and right. ⚡ headers and footers can span multiple lines. They can be defined through function calls or appropriately formatted data.frame with a template. Better! You can assign a footnote to a cell so a selected symbol (like *, **, ***, etc) will appear next to the cell content and appropriate entry will be added to the footer line(s). This way the "*" can be added to ALL occurrences of the content! ⚡ any kind of formatting can be conditionally applied to rows and columns or concrete sets of cells. No more hardcoded i=1, j=12 if your table is complex, you can employ regular expressions! ⚡ You can apply any kind of formatting: bold, italic, underline, superscript, subscript, colour, font name, font size. Markdown or HTML tags can be used through the 𝙛𝙩𝙀𝙭𝙩𝙧𝙖 package. Colours and other formatting is supported also natively, in a more structured way, and you can mix all of them through "paragraphs". ⚡ you can control the page layout (portrait, landscape), control margins, paddings in every cell. ⚡ Of course you can embed pictures, mini-graphs directly into cell tables, if you need this. ⚡ with the 𝙤𝙛𝙛𝙞𝙘𝙚𝙧 package you can read/create/manipulate MS Word documents (docx) and control dozens aspects of its formatting! You can use Word templates with predefined styles, text, chapters, numbered lists, placeholders for a content, and this way - reuse your corporate styles! No need to set fonts and colours manually every time - assign styles! ⚡ You can prepare the entire document with chapters, texts, introductions, etc. and appropriately styled placeholders for tables and figures and add some meta-language defining what should be put here. Then, you can find these placeholders, read assigned commands, run corresponding scripts, catch their output, and replace these placeholders with actual values! Yes, a document can play a role of a self-driving template-based analysis scenario! WOW! I only scratched the surface! ⛸ There's not enough space in a LI post to describe everything that David Gohel made himself for statistical programmers! You can save it as DOCx, PPTx, RTF, HTML! ⚡ tip: you can transfer your base R and ggplot2 plots into scalable "metafiles" (vector graphics), which can be resized without blurring, and embed it into your DOCx! 🔗 Flextable book: https://lnkd.in/deqNET8g 🔗 ftExtra: https://ftextra.atusy.net/ 🔗 officedown: https://lnkd.in/dgpuykXD + https://lnkd.in/dezAMw6v 🔗 officer: https://lnkd.in/gWc7sGA #sas #rprogramming #clinicalresearch #rstats #clinicalprogramming #statisticalprogramming #tflshell

89

16 Comments

Observational studies: practical tips for avoiding common statistical pitfalls 🌐 https://lnkd.in/g_NTHxKB The main objectives of statistics can be summarised as follows: 1️⃣ Describing data. Statistical analysis helps summarise and describe a dataset’s main features, such as central tendency (mean, median, mode), dispersion (range, variance, standard deviation), and distribution shape. 2️⃣ Inference and generalisation. It allows researchers to make inferences about a population based on a sample of data. It provides tools for estimating parameters and testing hypotheses about population characteristics. 3️⃣ Decision making. Statistical methods are used to support decision-making processes by providing a basis for drawing conclusions from data. This is particularly important in medicine, public health and social sciences. 4️⃣ Prediction. Statistical models can be used to predict future events or trends based on historical data. 5️⃣ Comparisons and relationships. Statistical analysis helps compare groups, study relationships between variables, and identify patterns in data. 6️⃣ Research design. Statistical analysis plays a vital role in designing experiments and surveys, helping researchers choose the appropriate data collection and analysis methods. 7️⃣ Policy formulation. In social sciences and government, statistical findings are often used to formulate and evaluate policies. It provides a basis for understanding social, economic, and demographic trends. 8️⃣ Risk assessment. Statistics assesses and manages public health risks, creates surveillance systems, and assesses potential threats. 9️⃣ Driving further research Statistical findings help to generate new hypotheses and drive further research.

17

Approaches to projecting future healthcare demand To learn how to implement #economic_evaluation techniques (#CEA and #CUA) using decision analytic models, including #decision_trees and #Markov_models in a one-to-one setting: https://lnkd.in/dqTpcvTM.

26

Different types of data need to be visualised and analysed in different ways. Whatever type of data your research is generating we have you covered. Book now to join biostatistician Katrina Gore and Dr Nikki Osborne LIVE online 7-9 October. For more information or to register visit https://lnkd.in/g3wvCKwp #researchtraining #CPD #ECR #PhD #statisticalanalysis

3

Hi all. I updated my review of free to use, easy to use, statistical software. The review now includes -gretl-. Free To Use Statistical Software: Comparing Statistical Analyses https://lnkd.in/er6CEMky I check to make sure that different packages all give the same results for frequency, means, correlation and OLS regression. Happily, they all do. In this review, I look at statistical packages that are free to use, and that are menu driven, so you don't need to learn coding. I hope you find this useful.

3

Introduction to Data Analysis with R (Data with R) With Examples from Public Health and Epidemiology https://datawithr.com/

7

Faster Analysis & Improved esolution for AAV Aggregate Analysis Exclusive access to online studies with useful information on AAV Aggregate Analysis. https://ow.ly/Osqo50RwaCn

1

In this video we demonstrate how to use the functions within the explore command in #SPSS #Statistics to test your data for normality. https://lnkd.in/eX4A-R6R

1