7. I.S. Wright and associates did a clinical trial on the effect of anticoagulan

Question

7. I.S. Wright and associates did a clinical trial on the effect of anticoagulant therapy for coronary heart disease. Eligible patients who were admitted to participating hospitals on odd days of the month were given the therapy; eligible patients admitted on even days were the controls. In total, there were 580 patients in the therapy group and 442 controls. An observer says,

Since the odd-even assignment to treatment or control is objective and impartial, it is just as good as tossing a coin.

Do you agree or disagree? EXPLAIN FULLY.Assume the trial was done in a month with 30 days.

Explanation / Answer

Solution:

I kind of agree with the observer. I can see where he is coming from, but I believe there is some important information missing. The randomization scheme is objective. As long as the patients are blinded to the randomization scheme, they should not be able to decide if they want to schedule their admittance on an odd or even day. However, you also need to also make the physicians administering the treatment blinded so that they, too, cannot pick and choose which patients get which treatment. If it is a double-blinded study, then this should work out to be the equivalent of flipping a coin (half the days are odd, and half the days are even).

However, if you were to take the distribution that turned out (580 vs. 442) and calculate the probability that you would get a 580/442 split, then you would see that that outcome is VERY unlikely. So, even though the idea does sound objective and should result in a fairly balanced distribution, it appears that there was some bias introduced (probably due to the lack of double blinding).

In short, I do not agree with the current information because the design does not speak of blinding thee randomization scheme.

Related Questions

Navigate

• Browse (All)
• Browse 7
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.