Clinical Trial Statistics: Sample Size, Randomisation & Survival Curves
Plan and analyse a fictional heart-failure trial: sample size and power, block randomisation, the baseline table, the primary endpoint with CI and p-value, NNT, a Kaplan-Meier curve by hand and interim boundaries, in a live spreadsheet that checks every cell. Educational, not medical advice.
The first lesson is free — no account needed.
About this course
A clinical trial statistics course that plans and analyses one fictional heart-failure trial in a spreadsheet, for students in medicine, nursing, pharmacy and public health, junior biostatisticians and anyone who reads trial papers. You calculate the sample size per arm for two proportions and two means with NORM.S.INV and inflate it for dropout, turn the formula around into the power of a smaller trial and chart a power curve, allocate patients with block randomisation and check the balance, build the baseline table with standardised differences, analyse the primary endpoint with a difference in proportions, a 95% confidence interval and a p-value, translate the result into the number needed to treat and to harm, build a Kaplan-Meier survival table and step chart by hand and test an interim look with O'Brien-Fleming and Bonferroni. Fictional data, educational only, not medical advice. The first lab is free.
Lessons
- Sample Size: Two Proportions and Two Means with NORM.S.INVFree
- Power for a Given Sample Size and a Power Curve Pro
- Block Randomisation: Allocating Patients and Checking Balance Pro
- The Baseline Table: Means, SDs and % by Arm Pro
- Primary Endpoint: Difference in Proportions, 95% CI and p-Value Pro
- Number Needed to Treat (NNT), Its CI and the Number Needed to Harm Pro
- Kaplan-Meier Survival Table by Hand + a Step Chart Pro
- Interim Analysis and Multiplicity: O'Brien-Fleming and Bonferroni Pro