Epidemiology question You conduct a randomized controlled trial to examine the e

Question

Epidemiology question

You conduct a randomized controlled trial to examine the effect of a new paleo diet to reduce obesity among clinic patients with a BMI > 30.Patients are randomly assigned to the new paleo diet or standard of care (low carb diet) as a comparison, and follow for six months. You notice during the follow-up period that 35% of the patients assigned to the standard of care (low carb diet) decided to try the paleo diet because they learned about it beneficial effects on Dr. Oz show.

How do you call this problem in your study? And what might be the effects on your study findings?

Explanation / Answer

This is a situation wherein 35% of subjects from control have moved from control to case during the course of observation.

From Statistical point of view, we can say that 35% of the control data is missclassified. (should have been in case but are in control )

So how to fix it?

There are two ways to fix it.

1. Add 35% to case from control and remeasure for shortened time period from point of transfer.

2. Cancel the 35 % from the study and make new contigency table comprising of data without missclassified data .

Effects on study findings

Let us try ot firstly quantify the effect of misclassified controls (i.e., individuals who have decided to try the paleo diet because they learned about it beneficial effects on Dr. Oz show ) on the ability of a case-control study to detect an association between a diet and obesity, and secondly, the effect of shifting misclassified controls in the study from control to case, as opposed to removing them (thus decreasing sample size). The results show that if, for example, 10% of controls are misclassified , the study’s informativeness is reduced to approximately 81% of what it would have been in a sample with no misclassified controls, whereas if these misclassified controls are removed from the study, the informativeness is only reduced to about 90%, despite the reduced sample size. If 25% are misclassified, those figures become approximately 56% and 75%, respectively. For 35 % missclasisfication ,the figues are 35% and 56% respectively. Thus, shifting the misclassified controls in the control sample is worse than removing them altogether.

Related Questions

Navigate

• Browse (All)
• Browse E
• Subjects

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