3. Who Gets Into a Trial: Eligibility, Run-In, Regression to Mean
After this lesson you can
- Write inclusion and exclusion criteria with a reason for every line
- Weigh homogeneity against generalizability
- Build a recruitment plan from a screening funnel and a monthly target
- Explain what a run-in period does and when to avoid one
- Recognize regression to the mean and explain why a control arm answers it
Video coming soon
Lessons publish daily from November 17, 2026.
This lesson covers who gets into a trial: the five populations from target to randomized, eligibility criteria with a reason for each, recruitment from a screening funnel and informed consent as a process. It explains what a run-in period buys and costs and shows regression to the mean in the fictional Oakhaven screening and baseline values.
Questions and answers
Questions a learner would ask about this lesson. Think through your own answer before you open each one.
- What are the five populations in a trial?
Show answer
The target population (whom the question is about), the accessible population (whom the trial can reach), the eligible (who meet the criteria), the consented (who agree to take part) and the randomized (who are in the trial). In the fictional Oakhaven funnel, 1,000 are assessed, 500 are eligible, 300 enter the run-in and 250 are randomized.
- What is the trade-off when eligibility criteria are made narrow?
Show answer
A narrow, homogeneous sample varies less, so the effect is easier to see with a smaller sample. The cost is generalizability: the result describes fewer people, and clinicians must apply it to everyone else. Randomization protects internal validity only, not external validity.
- What does a run-in period do, and what does it cost?
Show answer
A run-in is a short period between consent and randomization in which everyone receives the same thing and a rule decides who goes on. Oakhaven's two-week run-in requires attending two of three introductory sessions. It buys likely adherers and fewer dropouts, but the result then describes adherers only and may be smaller in ordinary clinics.
- Why did Oakhaven's mean control value fall from 72 to 66 before anyone was treated?
Show answer
That is regression to the mean. Participants were selected for a control value of 60 or above on one measurement, and a high value is partly true level and partly occasion noise. On the next measurement the noise is fresh, so the group mean drifts back toward the average, a fall of six with no treatment.
- Is informed consent just the signed form?
Show answer
No. Consent is a person's voluntary agreement based on understanding, and it is a process that begins before the form and continues after, with re-consent when the protocol changes. The lesson lists eight elements: purpose, procedures, risks, benefits, alternatives, voluntariness, confidentiality and contacts.
Practice quiz
- Question 1Of 1,000 people assessed in the Oakhaven funnel, how many were randomized?
- Question 2In the Oakhaven funnel, 500 were eligible and 300 entered the run-in. How many eligible people declined?
- Question 3Oakhaven needs 250 randomized participants over a ten-month recruitment window. What is the monthly target?
- Question 4A run-in period mainly improves:
- Question 5The mean control value fell from 72 at screening to 66 at baseline with no treatment. The best explanation is:
- Question 6Why does a control group answer regression to the mean?
- Question 7Making eligibility very narrow tends to:
- Question 8How does a run-in differ from eligibility screening?
- Question 9Which is an element of informed consent?
- Question 10Recruitment is falling behind the monthly line. Which response does the lesson rule out?
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.