• observational
  • analytic- hypothesis testing
  • descriptive- only describes
• experimental
  • analytic- always hypothesis testing

• Are “observing” what has occurred naturally
• Investigator does not manipulate exposure status of study participants
• Most epidemiologic studies are observational

• retrospective cohort
• prospective cohort
• nested case control
• case cohort

• case series
• case reports
• surveillance
• ecological
• descriptive survey
  • cross sectional

• Availability of data on exposures and disease
• Can be done quickly and with limited resources
• Exposures may differ substantially between cities, states, and countries
• Analysis is fairly simple

• Are subject to numerous biases and limitations
• Are more subject to confounding (the mixing of effects of other risk factors with the exposures of interest) than individual risk studies
• Time trend studies are limited in that you can’t determine whether exposure preceded the outcome
• Are subject to the “ecological fallacy”

• Results from correlations found on a group level do not hold true for the individual members of these groups.
• We don’t know the link between exposure and disease among individuals within each group

• Depends on the purpose
  • to simply describe the health status of the population (descriptive), or
  • to conduct comparative analyses, relate exposure to disease (analytical)

• prevalence odds ratio 
  • Estimates the Incidence Rate Ratio (IRR)
  • Best for X-sectional studies of chronic diseases
• prevalence ratio
  • Estimates the CIR
  • Best for X-sectional studies of acute diseases

• Selection bias is major problem
  • Diseases of long duration (e.g., diabetes) are often over-represented (“survival” bias)
  • Many individuals have left “cohort”
  • Representativeness of volunteers
• Time sequence between exposure and disease inferred only through history
• Often reply on subjective information

  Ex: Women exposed to pesticides in a study of breast cancer

   Ex: Cancer, cardiovascular disease

Ex: Studying risk factors for birth defects

Ex. Early studies of AIDS

Ex. Residence in a city

• Community-Based (Population-Based)
  • Prevalence case-control study
  • Incidence case-control study
• Cohort-Based (Nested)
• Case-Crossover Study

–Most often used when cases are selected from a defined geographic population

–Sources: random digit dialing, residence lists, drivers’ license records 

  Ex: Study of Hepatitis A in Hispanic Children; controls were chosen from same neighborhood and schools as cases

• Disadvantage 
  • Advantage
    • Because of selection process, investigator is usually assured that they come from the same base population as the cases.

–  time consuming

–  expensive

–  hard to contact and get cooperation

–  may remember exposures differently than cases (recall bias)

• Used most often when cases are selected from a hospital population
• Study of cigarette smoking and myocardial infarction among women. Cases identified from admissions to hospital coronary care units. Controls drawn from surgical, orthopedic, and medical unit of same hospital. Controls included patients with non-coronary conditions

•performed after cohort study

•choose all cases of disease among cohort

•identify sample of controls from among cohort members without the disease

•obtain detailed data on exposures and potential confounders

compare prevalence of exposure between cases and controls

(1) controls are representative of individual in the base population

(2) cases are representative of all individuals with the disease of interest in the base population

(3) the disease is relatively rare rare

• Useful for rare diseases
• Can assess multiple exposure & confounders
• Can assess exposures which are not detailed in records
• Smaller study population
• Quicker and cheaper