22 Chapter 1 Basic Ideas How can we prevent confounding? One way is to design a study so that the confounder isn’t a factor. For example, to determine whether smoking increases the risk of heart attack, we could compare a group of male smokers to a group of male nonsmokers, and a group of female smokers to a group of female nonsmokers. Gender wouldn’t be a confounder here, because there would be no differences in gender between the smoking and nonsmoking groups. Of course, there are other possible confounders. Smoking rates vary among ethnic groups, and rates of heart attacks do, too. If people in ethnic groups that are more susceptible to heart attacks are also more likely to smoke, then ethnicity becomes a confounder. This can be dealt with by comparing smokers of the same gender and ethnic group to nonsmokers of that gender and ethnic group. Designing observational studies that are relatively free of confounding is difficult. In practice, many studies must be conducted over a long period of time. In the case of smoking, this has been done, and we can be confident that smoking does indeed increase the risk of heart attack, along with other diseases. If you don’t smoke, you have a much better chance to live a long and healthy life. SUMMARY In an observational study, when there are differences in the outcomes among the treatment groups, it is often difficult to determine whether the differences are due to the treatments or to confounding. EXAMPLE 1.16 Determining the effect of confounding In a study of the effects of blood pressure on health, a large group of people of all ages were given regular blood pressure checkups for a period of one year. It was found that people with high blood pressure were more likely to develop cancer than people with lower blood pressure. Explain how this result might be due to confounding. Solution Age is a likely confounder. Older people tend to have higher blood pressure than younger people, and older people are more likely to get cancer than younger people. Therefore people with high blood pressure may have higher cancer rates than younger people, even though high blood pressure does not cause cancer. Check Your Understanding 1. To study the effect of air pollution on respiratory health, a group of people in a city with high levels of air pollution and another group in a rural area with low levels of pollution are examined to determine their lung capacity. Is this a randomized experiment or an observational study? 2. It is known that drinking alcohol increases the risk of contracting liver cancer. Assume that in an observational study, a group of smokers has a higher rate of liver cancer than a group of nonsmokers. Explain how this result might be due to confounding. Answers are on page 26. Objective 4 Describe various types of observational studies Types of Observational Studies There are two main types of observational studies: cohort studies and case-control studies. Cohort studies can be further divided into prospective, cross-sectional, and retrospective studies.
navidi_monk_essential_statistics_1e_ch1_3
To see the actual publication please follow the link above