Mini-project
Healthcare & Hospitals
Does the treatment work? A clinical trial analysis
Compare treatment and control groups in a public clinical trial with the right statistical tests.
The brief
Doctors read trial results in journals, but rarely see the analysis behind them. Take a public clinical trial dataset, check the groups were comparable, compare outcomes between treatment arms with hypothesis tests and confidence intervals, and present the result clearly. Good looks like: a balance check, correct tests for each outcome, effect sizes with uncertainty, and an honest summary.
What you will build
- Explore the trial data and its arms
- Check that baseline characteristics are balanced
- Compare outcomes with t-tests, chi-square and survival curves
- Report effects with confidence intervals
- Write a plain-language summary of the result and its limits
Free, open data to build it with
- AIDS Clinical Trials Group Study 175 ↗ UCI Machine Learning Repository
- Heart failure clinical records ↗ UCI Machine Learning Repository
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.