0. What a Clinical Trial Is, and Why We Randomize Participants
After this lesson you can
- Distinguish experimental from observational studies by who assigns the intervention
- Explain confounding by indication with a simple arrow diagram
- Describe what adjustment can and cannot fix in an observational comparison
- Explain balance in expectation and what randomization does not do
- Name the eight parts of a trial and the lesson that teaches each one
Video coming soon
Lessons publish daily from November 17, 2026.
The orientation lesson defines a clinical study and a randomized controlled trial and shows why an observational comparison of people who chose a program is threatened by confounding by indication. It explains how randomization answers that bias and introduces the fictional Oakhaven Trials Unit and the trial that runs through the course.
Questions and answers
Questions a learner would ask about this lesson. Think through your own answer before you open each one.
- What actually makes a study a trial rather than an observational study?
Show answer
The difference is who decides which person gets which intervention. In an experimental study the investigator assigns the intervention by a rule written before anyone enrolls. In an observational study the investigator only watches who got it. The setting, the size and the presence of a comparison group do not decide it.
- What is confounding by indication?
Show answer
It is the bias that arises when the reasons a person receives an intervention are themselves linked to the outcome. In the lesson's fictional example, people who join the support program may be more motivated, less ill or better supported at home, so a comparison of joiners and non-joiners mixes the program's effect with that head start.
- Can't adjustment fix an observational comparison?
Show answer
Adjustment can balance factors that were measured, such as age and the starting control value. It cannot balance reasons nobody measured, such as motivation, support at home or clinician judgment. Matching, restriction and natural experiments share the same limit, so the unmeasured head start remains.
- Does randomization guarantee that the two arms are balanced?
Show answer
No. Randomization gives balance in expectation, meaning on average over the trials that could have been run, for known and unknown factors alike. Any one trial can still be imbalanced by chance, though larger trials make chance imbalance smaller.
- What is the Oakhaven Adherence Trial?
Show answer
It is the fictional trial that runs through all sixteen lessons: a two-arm, parallel-group, individually randomized superiority trial of a twelve-month support program added to usual care versus usual care alone. The primary endpoint is poor control at twelve months, and it plans to randomize 250 adults, 125 per arm. Every number in it is invented for teaching.
Practice quiz
- Question 1What makes a clinical study experimental rather than observational?
- Question 2In the cold open, Clinician D notes that people who join the support program do better. Why can't that observation settle whether the program works?
- Question 3A randomized controlled trial is best described as a study that:
- Question 4Which statement about adjustment in an observational study is correct?
- Question 5"Balance in expectation" means that:
- Question 6Which of these does randomization NOT do?
- Question 7How does allocation concealment differ from blinding?
- Question 8The Oakhaven Adherence Trial plans to randomize how many adults in total?
- Question 9In the Oakhaven case, what counts as "poor control"?
- Question 10Where do the numbers used in this course come from?
Saved in your browser only: no account, no server.
Design of Clinical Studies is education only. Every lesson follows the fictional Oakhaven Adherence Trial with invented numbers; the course uses no real data and cites no real studies.