Students learn that the goal of epidemiology is to identify probable causes and prevent disease or other health problems. They also learn that it is often difficult to conduct experimental studies in epidemiology due to ethical and practical reasons. Students compare and contrast observational studies to experimental design studies. They learn that a case control study is one approach used to identify an association between a particular factor and health outcome in human subjects. Students learn the difference between association and causation. Finally, students learn about key concepts in case control studies (cases, controls, exposures, outcome, and matching)
Class time: 100 minutes (2 class days)
|Learn the components of the case control study design, including cases, controls, exposures, and outcome||Students are able to identify the cases, controls, exposures, and outcome for the smoking behavior study and other case control studies they identify through an online search|
|Understand the importance of matching cases and controls.||Students describe and discuss the potential consequences of not matching case and control subjects in the context of the smoking behavior study, such as drawing inaccurate conclusions about the smoking behavior of a specific group.|
Day 1: Case control study and introduction to causality
Section A. Conducting Research on Human Subjects- Introduction to epidemiology with the Beta Frat House party scenario
|Video- Reviewing the Beta Frat house case control study
1. Present students with the following scenario:
“The Beta Fraternity House held a party on Friday night. By 4 pm on Saturday afternoon, 17 people who had been at the party had called Doug, the frat house manager, complaining of headaches and severe nausea. None of the frat brothers who were absent from the party reported feeling sick.”
2. Ask students what Doug should do to figure out what was making people sick? In other words, ask students to propose how they would determine the cause of the sickness.
Students will probably suggest doing a comparison of the people who got sick and those who didn’t. Encourage them to suggest what factors might have caused the sickness, as well as how they would go about gathering information. Students will probably recognize that they need to look for one or more factors that are more common among the people who got sick than those who didn’t get sick.
3. Tell students that epidemiology is a field of science that investigates disease in populations and tries to identify probable causes. There are several kinds of studies frequently conducted in epidemiology. For an investigation like this one, in which an outbreak of a sickness has occurred, epidemiologists would often use a case control study. It may be helpful to hand out Epidemiology Glossary for reference.
4. Continue with the Beta Fraternity House party:
“Doug reported the outbreak to Student Health Services. The public health team there designed a case control study to try to find out what had made the students sick. Their study design consisted of giving a questionnaire and a health exam to as many students as they could locate who had been to the party, including those who got sick and those who didn’t. Both groups of students in the study—those that got sick and those that didn’t—included males and females in the age range of 18 to 22. The questionnaire asked students about what foods they had eaten, what beverages they had consumed, and how much alcohol they drank. It also asked them to list the people they had been in contact with during the party. During the health exam, the nurse measured the students’ temperature, blood pressure, and other vital signs, as well as asking them how they felt. The students who took the questionnaire were asked to report back to Student Health Services one week after the party for a follow-up health exam. Results from the study were analyzed to see whether there were any differences between the students who got sick and those who didn’t.”
5. Complete the flowchart in Blank_Student Sheet 4.1 as a class using the above example to demonstrate the characteristics of a case control study, as shown in Completed_Student Sheet 4.1. Hand out Epidemiology Glossary and define epidemiology, cases, controls, study sample, outcome, and exposures when going through Student Sheet 4.1.
6. Ask students whether they can suggest other examples where a case control study design might be an effective approach for determining what might have caused the outcome being investigated. (They may suggest specific cancers or outbreaks of food poisoning or infection. They should recognize that the outcome has already occurred, and the study would be looking backward in time to analyze what factors might have contributed to the outcome.) You may want to draw the analogy between a case control study and solving a crime—in both, the outcome has occurred, and the investigator is faced with the challenge of determining what might have caused it.
|Comparing experimental design studies to case control studiesSometimes students are confused by the differences in design and terminology between case control studies and the classical experimental study usually presented as the “gold standard” in high school science classes. The activity presented in Blank_Student Sheet 4.2 is intended to help students understand the similarities and differences. This activity describes two research studies, the case control study just discussed and a classical experimental study. Students analyze the two studies and show the similarities and differences between the two by filling in a table.
As students complete Student Sheet 4.2, they should be able to see the parallels between experimental and case control studies. For example, they should recognize that: 1) matching cases and controls is analogous to having controlled variables in a classical experiment; 2) an exposure is similar to a manipulated variable; and 3) the outcome of a case control study is analogous to the responding variable in the experimental study.
They should also note the differences between the two designs, including: 1) an experimental study is prospective (looks forward in time), while a case control study is retrospective (looks backward in time); 2) The responding variable is measured toward the end of the classical experiment, while the analogous outcome is present in case subjects at the start of a case control study; 3) the researcher deliberately chooses the manipulated variable in an experimental study, whereas the exposures have occurred independent of the researcher in a case control study; and 4) in a carefully controlled experimental study, we can usually say that the manipulated variable causes the responding variable, while in a case control study, we say that the exposure is associated with the outcome. Students will learn and apply the criteria for causation in Lesson 5.
|Video- Class discussion of matching and comparison of case control studies and experimental design studies
7. Optional Ethical Consideration discussion:
Ask your students to discuss the following question: “When investigating whether a particular factor causes an outcome like a sickness in humans, why do researchers often use a case control study instead of an experimental study?”
Students should recognize that it is unethical to deliberately expose human subjects to a factor that might make them sick. You might ask students to suggest or research examples where experimental protocols were conducted on human subjects that lead to harm to the subjects.
For more information on conducting discussions of bioethics in your classrooms, refer to An Ethics Primer, a definitive resource for teaching this topic (Chowning, J. & Fraser, P. (2006). An Ethics Primer. Seattle WA: Northwest Association of Biomedical Research.http://nwabr.org/curriculum/ethics-primer).
Section B. Causality vs. Association
1. After completing Student Sheet 4.2 students should understand that observational studies by themselves, such as case control studies, can only show an association whereas an experimental design study can show causality.
2. Explain that there is a difference between causality and association. For example, people with red hair often have freckles, so they are definitely associated. But red hair does not necessarily cause freckles. Ask students if they can think of any other examples where two things are associated but one doesn’t cause the other.
|Epidemiologists detect associations through observational studies, but need to go beyond the results of one observational study to infer causality. Explain to students that epidemiologists define a cause as an exposure which increases the risk of someone developing an outcome. Causality interpretation diagram may be helpful in understanding the causality definition.|
3. Tell students that when two things (A and B) are associated, there are three possible explanations for the association. Use Figure 4.2- Causal Diagram as you explain the causal diagram pathway. Tell students that for an exposure to cause an outcome, the directional arrow must point to the outcome and originate from the exposure.
4. Refer to the Figure 4.3- Criteria for Causality with your class. (In lesson 5, students will work with the Criteria for Causality in an example question from the smoking behavior database and will apply the criteria to their research hypothesis in lesson 6.). Explain to students that when analyzing case-control studies, epidemiologists need to apply a number of criteria to a detected association to help determine whether the exposure is likely to have increased the risk of developing the outcome. The more criteria that are satisfied, the more confident one can be that the association is causal, however the process is somewhat subjective with consensus built after many peer-reviewed studies.
Day 2: Learning about the Smoking Behavior Study and introduction to matching
Section C. Student sheets: Using the questionnaire to test a research hypothesis
1. Ask students to get out their questionnaires, and give them Blank_Student Sheet 4.3.
2. Bring up the database (https://gsoutreach.gs.washington.edu/database2) on your projector – Step 1.1 Subject-classification and Matching Questions
3. Explain to students that the different questions in the research questionnaire have specific purposes. Below is a table that shows the specific purposes of the questions. As you carry out a short review of these, ask the students to color code their questionnaires by circling question numbers, using the color code in the table that follows. The database is also color coded in this fashion.
Section D. Research project pages: The smoking behavior research study
1. Make sure students have a copy of Blank_Research Project Pages Lesson 4. They can fill in information about the study as you discuss it with them.
2. Explain to the students that they’re going to learn more about the design of the smoking behavior research study through a discussion and by looking at the data in Step 1.1 Subject-classification and Matching Questions.
3. Show students Figure 4.1- Stages of Smoking, the diagram of the stages of smoking that shows who the cases and controls are. They saw a similar diagram in Lesson 2. In addition, guide the class in examining the data in Step 1.1 Subject-classification and Matching Questions as they complete the research page for this lesson. You may also want to show the students the video below.
4. Ask students to fill in the data on Blank_Research Project Pages Lesson 4 or give students Completed_Research Project Pages Lesson 4 and discuss the differences between cases and controls for these questions.
|Background on the Smoking Behavior Research StudyFacts about the Smoking Behavior Research Study
HOMEWORK (optional): A quick internet search will show students how common case control studies are.
Ask them to go to http://scholar.google.com/ and search on the term “a case control study.” Provide them with a copy of Student Sheet 4.1, and ask them to complete this sheet for one of the studies that they find in their search.