Question

13. A group of scientists decides to conduct a study addressing the long-term effects of maternal alcohol consumption on their infants after conception. Two hundred women, including those who suffer from alcoholism and those who do not, are recruited into the study when they present for their first primary care visit. A medical history is taken on alcohol use, prena- tal care, nutritional status, and smoking behav- iors; these are measured monthly during the pregnancy. The researchers follow the wom- en’s pregnancies until term, after which they devote their attention to the health and behav- iors of the offspring. Which of the following is the most appropriate statistic the research- ers will be able to calculate as a result of their study?

1. (A)  Attributable risk of offspring abnormalities in mothers who smoke

2. (B)  Odds ratio of offspring abnormalities in mothers who consume alcohol during pregnancy

3. (C)  Prevalence of alcohol consumption during pregnancy

4. (D)  Proportion of all offspring abnormalities that are due to alcohol consumption dur- ing pregnancy

5. (E)  Relative risk of offspring abnormalities in mothers who consume alcohol during pregnancy

Answer 1

The correct answer is E. The study described here is a cohort study, because it includes a group with and a group without a given risk factor (fetal exposure to alcohol) and then looks at whether the risk factor changes the chances of offspring getting the disease (abnor- malities). The study is prospective, because the group members are looked at before the dis- ease (abnormality) develops in the offspring. Relative risk can be calculated from the results of a cohort study by comparing the rate of dis- ease in the group with the risk factor to the rate of disease in the group without the risk factor.

Answer A is incorrect. Attributable risk can be calculated from the results of a cohort study and describes the proportion of disease that is due to the risk factor under study. Although smoking behavior of the women is being re- corded, the study is not designed to look at the impact of this risk factor on fetal abnor- mality; the rate of smoking in the two groups of women is unknown, and thus we do not know whether there are sufficient numbers of women in the “exposed” and “unexposed” groups when it comes to tobacco.

Answer B is incorrect. An odds ratio is similar to relative risk, but it is calculated from the re- sults of a case-control study, not from a cohort study. Because birth abnormality is a relatively rare outcome, the odds ratio from a case-control study would likely closely approximate the ac- tual relative risk.

Answer C is incorrect. Prevalence is a mea- sure of how many cases of a given disease exist in a population that is at risk for that disease. This is not the best answer in this case, be- cause there are no data to judge whether this group of 200 women represents the true preva- lence of alcoholism during pregnancy in the community.

Answer D is incorrect. The statistic described in this answer is the population attributable risk (PAR), which helps us understand, in a given population, how much less common a disease (fetal abnormality) would be if a given risk factor (alcohol consumption during preg- nancy) were completely eliminated. To calcu- late PAR, the attributable risk is multiplied by the prevalence of the risk factor in the popula- tion, which would not be known in this study.