Case-control studies
If you want to find out more about health conditions in your village it would be necessary to use a slightly more complicated research design than a cross-sectional study. Usually, this will involve a case-control study. You begin by selecting two groups for comparison: one group of the population (the cases) who have a characteristic that you wish to investigate (such as mothers whose child died within one month of birth, or children with malnutrition), and another group (the controls) where that problem is absent. The aim of a case-control study is to find out what factors have contributed to the problem under investigation. Case Study 14.2 gives an example to illustrate this type of study.
Case Study 14.2 A case-control study of the causes of newborn deaths
If you want to study the causes of neonatal (newborn) deaths in your area, you would first select the cases (babies who died within the first month of life) and the controls (babies who survived their first month of life). Then you would interview their mothers to compare the history of these two groups of babies, to determine whether certain factors are more common among the babies who died than among those who survived.
What feature of Case Study 14.2 indicates that this is a case-control study?
Two groups were compared to see if there were any differences in their situation that could explain why the cases died within the first month of life and the controls survived the first month. So this is a case-control study.
It is important to ensure that cases and controls in your study come from the same study population. For example, in a case-control study of vaccination of pregnant women against tetanus, where cases of unvaccinated women are being selected from a particular health centre, controls should be selected from mothers who have attended vaccination at the same health centre. If controls are selected from another centre, they would not be from the same study population and therefore they would not be directly comparable to the cases.