Epidemiology Assignment, Coursework Example

Consider each of the following scenarios and explain whether the variable in question is a confounder:

A study of the relationship between exercise and heart attacks that is conducted among women who do not smoke. Explain whether gender is a confounder.

In this example, gender is not a confounder.  There are two ways to reach this result.  First, the study only contains women; thus, there is no possibility of confounding between men and women. Second, it’s not clear that even in that case, that gender would be associated with the exposure (exercise) and heart attacks (outcome).

A case-control study of the relationship between liver cirrhosis and alcohol use. In this study, smoking is associated with drinking alcohol and is a risk factor for liver cirrhosis among both non-alcoholics and alcoholics. Explain whether smoking is a confounder.

In this example, smoking could be a potential confounder.  That, is smoking satisfies the three criteria of a confounder: 1) related to the exposure(drinking alcohol) ; 2) related to the outcome (liver cirrhosis); 3) not on the causal pathway between alcohol use and liver cirrhosis.

Interpret the results of the following studies

An odds ratio of 1.2 (95% confidence interval: 0.8-1.5) is found for the association of low socioeconomic status and occurrence of obesity.

The odds ratio in this example does not indicate association between the exposure (low socioeconomic status) and the result (occurrence of obesity). That is even though the odds ratio, not looking at the confidence interval, indicates association.  However, because the confidence interval (0.8-1.5) includes 1 (the null); thus,  one cannot conclude there is an association between the two variables.

A relative risk of 3.0 is reported for the association between consumption of red meat and the occurrence of colon cancer. The p-value of the association is 0.15.

In this example, the relative risk for the exposure and outcome variables is 3.0.  This would lead one to believe that there may be an association between consumption of red mean and occurrence of colon cancer.  However, with a p-value of .15- this evidence leads one to believe that the association is not statistically significant.  Thus, one would probably want to conduct additional analysis in order to understand if there is confounding or effect modification in this particular example.

An odds ratio of 7 (95% confidence interval: 3.0 – 11.4) is found for the association of smoking and lung cancer.

The odds ratio in this example (7.0) indicates a fairly robust association between the exposure (smoking) and the outcome (lung cancer).  Although the strength of this association varies widely based on the confidence interval (from 3.0-11.4), this may be a function of sample size.

The relationship between cigarette smoking and lung cancer was conducted in a case-control study with 700 cases and 425 controls. Using the results below, calculate the crude odds ratio and explain what the ratio means:

• Heavy Smoking—Cases: 450; Controls: 200Not Heavy Smoking—Cases: 250; Controls: 225
• Crude Odds ratio: (450*225)/ (200*250)= 2.03

From this crude odds ratio (2.03), and without knowledge of the confidence interval, there is a putative association between the level of smoking (heavy smoking) and lung cancer.

A case-control study looked at the association of alcohol use with the occurrence of coronary heart disease (CHD). There were 300 participants in the study (150 cases and 150 controls). Of the cases, 90 participants drank alcohol; of the controls, 60 participants drank alcohol.

Design the appropriate 2×2 table, calculate and interpret the appropriate measure of association.
Crude Odds Ratio: (90*90)/ (60*60)= 8100/ 3600= 2.25

From this crude odds ratio (2.25), and without knowledge of the confidence interval, there is a putative association between consumption of alcohol (exposure) and occurrence of coronary heart disease (outcome).

You suspect that the association between alcohol use and CHD might be confounded by smoking. You collect the following data:

Calculate the appropriate measure of association between alcohol use and CHD in both smokers and non-smokers. Discuss whether smoking was a confounder of the association. What is the relationship of alcohol use to CHD after controlling for confounding?

In order to explore potential effects of confounding, one should calculate stratum-specific odds ratios on the stratifying variable (smoking), and then compare those results to the original odds ration

1. A) Smoking Odds Stratum Ratio- (80*10)/ (40*20)= 1
2. B) Non-smoking Odds Stratum Ratio (10*80) / (20*40)= 1

We have two stratum specific odds ratios (smoking and non smoking) of 1 indicating no association, and a crude odds ratio of 2.25.  Thus, one would likely conclude that due to the difference between odds ratios, smoking is indeed a confounding factor in the relationship between alcohol and coronary heart disease.

A study was conducted in young adults to look at the association between taking a driver’s education class (exposure) and the risk of being in an automobile accident (outcome). 450 participants were included in the study, 150 cases who had been in an accident and 300 controls who had not. Of the 150 cases, 70 reported having taken a driver’s education class. Of the 300 controls, 170 reported having taken a driver’s education class.

Calculate and interpret the appropriate measure of association between driver’s education and accidents.

Crude odds ratio: (70*130)/ (170*80)= .67

From this crude odds ratio (.67), and without knowledge of the confidence interval, there is a protective effect between the exposure and outcome.

The question arose as to whether gender might be an effect modifier of this association. When gender was assessed, the data looked like the following:

Perform the appropriate calculations to test for effect modification. Interpret your results.

In order to see if there is effect modification, the relationship between driver’s education training and accidents will be stratified by gender.

Women- (10*50)/ (50*40) = 500/ 2000= .25

Men-   (60*80)/ (120*40) = 4800/ 4800= 1.0

Looking at the results of the stratified analysis, there is likely effect modification in this example.  That is because the odds ratio between women and men is quite different indicating a differential effect based on gender.