Question

1. An epidemiologist was interested in determining whether use of non-steroidal anti-inflammatory drugs (NSAIDS) causes gastrointestinal bleeding. S/he recruited 600 subjects who had been hospitalized with gastrointestinal bleeding and 600 subjects from the local community who had never been diagnosed with such a bleeding problem. Due to concern over their diagnosis, subjects who had experienced a gastrointestinal bleed were able to provide a more accurate recollection of their NSAID use, while the comparison subjects were more likely to underestimate their use. Which design best describes this epidemiologic study?

1. quasi-experimental study
2. cohort study
3. cross-sectional study
4. case-control study

2. Which of the following best represents the bias described in this situation?

1. recall bias
2. confounding
3. selection bias
4. low sample size bias

3. If 300 of the subjects with gastrointestinal bleeding and 100 of the subjects without gastrointestinal bleeding reported using NSAIDS during their interview, what would have been the measure of association? (Hint: pay attention to the appropriate measure of association for this study design)

1. 1.5                                                         
2. 3
3. 5
4. 7

Answer 1

SOLUTION

1.case control study;Case control study is an epidemiological study that compares two groups of people that is those with the disease or condition under the study(cases) and a very similar group of people who do not have the disease condition(control).Here in this study epidemiologist using population with gastrointestinal bleeding and population with out gastrointestinal bleeding.This type of study has 3 distinct features.

• BOTH EXPOSURE AND OUTCOME HAVE OCCURED BEFORE THE START OF THE STUDY
• THE STUDY PROCEEDS BACKWARDS FROM EFFECT TO CAUSE
• IT USES CONTROL OR COMPARISON GROUP TO SUPPORT OR REFUTE AN INFERENCE.

2.Selection bias

In a case control study selection bias occurs when subjects for the control group are not truly representative of the population that produced the cases.The exposure distribution in cases is then compared to the exposure distribution in controls in order to compute odds ratio as a measure of association.

3.Here the answer is 5. Measure of association usually calculated using odds ratio.In case control study odds ratio can be calculated by the following equation.

	CASE( PATIENT WITH GI BLEEDING)	CONTROL(CASES WITH OUT GI BLEEDING)	TOTAL
EXPOSED TO NSAID	a(300)	b(100)	400
NON EXPOSED NSAID	c(300)	d(500)	800
total	a+c=600	b+d(600)	1200

odds of exposure(cases) =no.of cases with exposure =a/c=300/300=1

no.of cases without exposure

odds of exposure(control)= no.of controls with exposure =b/d =100=500=0.2

no.of controls with out exposure

odds ratio=ad/bd =300*500/100*300=5