Chapter 11 Review Questions Key

§11.1 - §11.3 Ecological & Cross-Sectional Studies

  1. This is an aggregate-level measurement.
  2. Aggregation bias occurs when an association seen in aggregate data does apply to the individual. 
  3. The relationship was highly confounded. Persons at low elevation were more likely to draw their water from the polluted sources.
  4. (1) contextual variable  (2) integral variable (3) contagion variable
  5. Here's one hypothesis: Larger hospitals may have more complete reporting..
  6. False. Longitudinal-ness depends on whether you can place events in individuals over time. (The temporal relation between occurrence of events and when they are measured is called proximity.)
  7. Notiones vulgares are crude unscientific personal beliefs natural and social phenomena.
  8. High SES people with mental disorders may be more likely to be diagnosed as neurotic; low SES people with mental disorder may be more likely to be diagnosed as psychotic. Also...differences in reporting mechanisms (e.g., legal authorities vs. private providers).
  9. (1) prevalence-incidence bias (2) detection bias (3) reverse-causality bias (cart-before-the-horse bias)

§11.4 Cohort Studies

  1. There is no long-term follow-up of individuals over time in the open population.
  2. TB's long latent period required long-term  follow-up of individuals, as is case in studies of chronic diseases.
  3. Framingham is a town in Massachusetts. We care because it is the seat of an innovative cohort study that has yielded important findings about heart disease. 
  4. These are the two investigators primarily responsible for the British Doctors study.
  5. Data for prospective studies occur concurrently to the investigation. Retrospective cohort studies are based on historical (nonconcurrent) information. 

§11.5 Case-Control Studies

  1. By studying only a subset of non-cases.
  2. There will be about 10 cases per year. Therefore, it will take 10 years to accrue 100  cases.
  3. The odds ratio
  4. The frequency of exposure in non-cases. 
  5. (c) rate ratio 
  6. 1
  7. True.