OBSERVATIONAL STUDIES

A. Observational Studies

1. Definition

   An OBSERVATIONAL STUDY is an assessment of treatments, policies, or exposures and their associated outcomes.

   Studies on the effects of smoking are always observational. See text p.12-13

   They differ from experiments in that the researcher has no control over which subjects are assigned the treatment. He or she collects data as they currently are. These study are common in fields that examine the effect of policies on people.

   Surveys are a typical example of observational studies. See the Breast Cancer handout or the Gallup Voter Poll handout.

2. Remarks

   a. Observational studies are useful for descriptive statements of fact. Example from the handout: white women who live in the San Francisco Bay Area have the highest breast cancer rates in the whole world.

   b. Observational studies are dangerous to use for cause-and-effect conclusions because of CONFOUNDING FACTORS (outside factors which can interfere with outcomes).

   Can't tell if it is something about San Francisco or something about the women who choose to live there that leads to the highest breast cancer rates in the world.

   c. Some confounding factors can be eliminated by CONTROLLING FOR that factor by looking at certain subgroups.

3. Other Examples (not discussed in lecture)

   a. According to actuarial records, people who drive red cars are more likely to get into accidents than people who drive white cars.

   (This is an observational study; as a statement of fact, it is true that drivers of red cars are more likely to be in accidents; but painting a person's red car white will not make them a better driver.)

   b. According to medical records, the disease pellagra was associated with flies: where there were flies, there was more likely to be pellagra.

   (Again, this is observational; flies are a useful marker of pellagra; but flies are also an indicator of poverty, and pellagra is a disease of malnutrition, not infection.)

   c. According to medical records, smokers are more likely to die of heart attacks than nonsmokers. A possible confounding factor is perhaps gender: men are more likely to smoke, and men are naturally at a higher risk of heart attacks. How can gender be eliminated as a confounding factor?

   (If only men are examined, and male smokers are compared to male nonsmokers, then gender plays no role, since everybody is male.)

   d. (Fisher's "constitutional hypothesis":) Suppose a researcher believes that smoking is confounded by genetics; some people have a gene within them that makes them want to smoke; and whether or not they smoke, the gene causes them to get lung cancer. How can genetics be eliminated as a confounding factor?

   (Compare monozygotic identical twins!)

C. Comparing Experimental and Observational Studies

1. Randomized, controlled experiments are more expensive and more difficult to do than observational studies.
2. Experiments can also be unrealistic (in an artificial setting) and unethical (smoking studies).
3. Controlled Experiments are better than observational studies in that a researcher can begin to eliminate confounding and pin down cause and effect. Here researchers impose a treatment on randomized subjects. This is not true of observational studies.

D. Vocabulary and Comments

• Observational Study
• Confounding Factors
• Controlling For
• Random Assignment
• Treatment Group
• Control Group
• Placebo, Placebo Effect
• Blind
• Double Blind
• Method of Comparison
• Replication, Replicate