HSEM 3708: CLA Seminar in Clinical Trials - Review Questions


School of Public Health


Division of Biostatistics


1. What is meant by the phrase 'intent to treat'? Why is it an important principle in clinical trials?

2. What is meant by 'surrogate endpoints' ? Explain the significance of this phrase in clinical trials.

3. Describe what is meant by Phase I, Phase II, and Phase III studies.

4. Why are clinical trials regarded as more reliable than case-control studies?

5. What are the key ingredients in estimating sample size for a clinical trial?

6. An investigator decides that randomization of patients is a nuisance. He decides to simply assign patients who appear on Monday, Wednesday and Friday to Drug A and those who appear on Tuesday, Thursday, and Saturday to Drug B. (No one appears on Sunday.) What might be wrong with such a scheme?

7. List 5 essential ingredients in an informed-consent document.

8. What are the stages in development of a new drug?

9. In your view, what mistakes (if any) were made in the testing of TGN1412 ?

10. What is a p-value?

11. What is statistical power?

12. An investigator wants to study the effects of a certain kind of eyedrops. The primary endpoint of the study is the change in the pressure within the eye (intraocular pressure). The investigator plans to randomize one eye of each participant to either the drug or a placebo, and the other eye will get the opposite. Is this a good design? Or might it be better to randomize each participant to get either the drug in both eyes or the placebo in both eyes? What are advantages and disadvantages of each of these two designs?

13. What was the importance of the main result of the ALL-HAT study?

14. What was the importance of the main result of the SMART study?

15. Why do you think some members of the Monitoring Board of the SMART study did not favor stopping the clinical trial at the time it was stopped?

16. Do you think the FDA is influenced by financial pressures? Other pressures? Describe.

17. Speculate on why Roger Poisson falsified eligibility data for the Breast Cancer Adjuvant Study.

18. How would you design a clinical trial of a new vaccine to prevent avian flu? What are the ethical considerations?

19. An investigator carries out a case-control study in which the cases are people who have had a heart attack and the controls are people who have not had a heart attack and who are the same age, gender, and smoking status as the cases. The investigator wants to see whether eating bacon, cheese or Caesar dressing, smoking, scuba-diving, wearing nylon socks, exposure to high- voltage lines, or use of Vicks Vapo-Rub are associated with heart attack. He carries out comparisons of cases and controls and finds the following: Factor p-value ---------------------- ------- Bacon 0.456 Cheese 0.988 Caesar dressing 0.067 Smoking 0.889 Scuba-diving 0.002 Nylon socks 0.439 High-voltage lines 0.771 Vicks Vapo-Rub 0.044 He notes that Scuba-diving and Vick Vapo-Rub seem to have low p-values. He concludes that these two are risk factors for heart attack, but that all the other things he investigated are not. Do you agree? Speculate on weaknesses of this study.

[More to be added ...]

Web address of this page: http://www.biostat.umn.edu/~john-c/HSEM3010.s2007.html

Most recent update: January 25, 2009