Background Story: One day, I need to validate a SPM related files and ADSPM in a hematology study, I communicated with an experienced colleague many times and finally got the work done. Since each study has different data structure, there is no standard SPM process available for all. I will share my experience with SPM work, hopefully it is helpful…

# Category: Clinical Trial

## Type I error vs Type II error

Background story: Last time, when we used R to calculate the sample size, we specified Type I error α and Type II error β, but what does the meaning behind α and β? Review: We define the “best” sample size that has less variation of the sample mean from sample to sample.

## One Factor Designs

Factorial Design allows the investigation of sets of categorical predictors, and the interaction between them. Today we will go over some basics of One Fixed Factor and One Random Factor Design. Keywords: Factor: categorical predictors Fixed Factors: Estimate the difference in means between groups defined by specified categorical predictors. e.x. ANOVA model, one measurement per subject. Random Factors: Estimate the variance…

## Non-inferiority Trial

Background Movie: Based on my taste, 我不是药神 Drug Deal/Dying to Survive is one of the best movies in 2018. It was based on a true story that some cancer patients use a kind of cheap Indian medicine (Veenat 100) as a substitution of the authentic Novartis Glivec 400 to prolong their lives. Besides the touching story, Veenat 100 is a…

## Interim Analysis

Background Story: Once, we need to do an Interim Analysis, I didn’t understand why we need to do it. Later I learned in Clinical Trial studies, our ultimate goal is trying to get approval for FDA submission at the end of studies. While exhausting and long progress, sometimes we perform Interim Analysis before the completed trial to access Safety, Futility,…

## Intention to Treat vs Per-protocol Analysis

Background Story: We often use different populations (Safety, Efficacy, ITT, Per-Protocol) to generate Clinical Trial Reports, but what are the differences between the populations? Today we will go over the basics of Intention to Treat and Per-protocol Analysis.

## Open Cdisc Report

Background Story: Once my boss asked me to review an Open CDISC report for SDTM dataset packages. I wasn’t sure what to do at first. With the help of my colleagues, I gradually develop a sense of how to review the CDISC report. I’d like to share some review processes and common Open CDISC Summary from PINNACLE21.

## Variable Types in Statistics

In data science, we usually classify variables into two big types: Continuous(age), Categorical(sex). What about the variables in clinical trials/statistics? How many different types do we have? Little goal: Today we will learn the six common types of variable in statistics? How to use and present them in clinical trials?

## One-tailed vs Two-tailed Test for P value

R day 3 In clinical trial studies, P-value is critical for measuring drug efficacy. Typically, the statistical analysis Plan will describe statistical significance as: P-value <0.05 for two-tailed Log Rank Test. Recently, I came across a study with the statistical significance as P-value <0.025 for one-tailed Log Rank test in the analysis plan. Question: Does two-tailed P-value= 0.05 means the…

## Restricted Mean Survival Analysis

Inspired by Joe Wang and Tony Qi, i did a summarization of Survival analysis and Restricted Mean Survival Analysis.

Restricted Mean Survival Analysis