Question

Suppose that 2000 pregnant women were identified during their 1^st trimester and completed questionnaires about health behaviors, including alcohol consumption. Among the 800 women who drank alcohol during pregnancy, 200 delivered low birth weight (LBW) infants. Among the 1200 who did not drink alcohol during pregnancy, 150 delivered low birth weight babies.

Complete the 2x2 table, labeling the exposure and disease boxes with the proper information. Show all your work for the calculations.

	Disease: Low birth weight babies	No Disease: Average birth weight babies
Exposure: Alcohol consumption
No exposure: No alcohol

What type of study design is this?

What is the appropriate measure of association?

Calculate the appropriate measure of association.

Can you calculate the AR% from the information provided? If so, calculate it and explain what it means.

Answer 1

The given is a case-control study.

Note that, the researchers did not determine whether the pregnant women consumed alcohol or not. They simply made an observational study, considering two comparable populations of pregnant women- those who consumed alcohol, and those who did not. Then, they observed the birth-weights of each of their babies, and attempted to identify whether the two populations varied with respect to the birth-weights of the babies (Disease: Low birth weight, which are the cases, and No Disease: Average birth weight, which are the controls).

Hence, it is a case-control study.

The table is completed below:

The odds ratio is an appropriate measure of association in case of a case-control study.

The odds ratio (OR) is calculated as follows:

OR

= [odds of exposure among cases] / [odds of exposure among controls]

= [(number of cases that had exposure) / (number of cases that had no exposure)] / [(number of controls that had exposure) / (number of controls that had no exposure)]

= [200/150] / [600/1,050]

≈ 2.33.

The value of the measure of association, odds ratio is 2.33.

Since the odds ratio is greater than 1, it can be said that the odds of alcohol consumption among women during pregnancy (exposure) is higher among women with babies with low birth weight (cases), than among women with babies with average birth weight. Hence, alcohol consumption might be a factor lowering the birth weight.

The AR% is the attack rate percent.

The AR% is calculated as follows:

AR%

= [(number of diseased cases in a group) / (total number of people in that group)] * 100%.

Group- Exposure: Alcohol consumption:

AR%

= [(number of low birth weight cases among women who consumed alcohol during pregnancy) / (total number of women who consumed alcohol during pregnancy)] * 100%

= (200/800) * 100%

= 25%.

Group- No exposure: No alcohol:

AR%

= [(number of low birth weight cases among women who did not consume alcohol during pregnancy) / (total number of women who did not consume alcohol during pregnancy)] * 100%

= (150/1,200) * 100%

= 12.5%.

Thus, among women who consumed alcohol during pregnancy, the AR% is 25%; among women who did not consume alcohol during pregnancy, AR% is 12.5%.

It can be observed that: (25%) / (12.5%) = 2. This means that AR% of the exposure group is 2 times that of the no-exposure group.

Thus, it can be said that women who consumed alcohol during pregnancy were 2 times more likely to give birth to babies with low birth weights, than women who did not consume alcohol during pregnancy.