1. EXPLANATORY
   1. Observational Studies: No Interventions
      1. Types
         1. Follow_Up Study
            1. Advantages
            2. Strongest observational study
            3. Permits direct determination of risk
            4. Easier, cheaper than clinical trials
            5. Exposure known at beginning of study
            6. Allows examination of multiple outcomes
            7. Components
            8. Sources of Cohorts
            9. population based
            10. exposure based
            11. Selection of comparison groups
            12. internal comparison
            13. 2x2 Table of Internal Comparison
            14. separate cohort control
            15. 2x2 Table of Separate Cohort Control
            16. comparison with availabe population rates
            17. Matching
            18. Exposed and unexposed individuals are similar in terms of demographics
            19. Used to ensure susceptibility to disease is equal except for presence of risk factor under investigation
            20. Alternative Names
            21. cohort
            22. prospective
            23. incidence
            24. longitudinal
            25. Disadvantages
            26. difficult to address rare events
            27. results not available for a long time
            28. expensive, but cheaper than clinical trials
            29. ethical concerns
            30. Concerns and Biases
            31. volunteer bias
            32. healthy worker bias
            33. migration bias
            34. surveillance bias
            35. non-response or loss to follow-up
            36. data analysis bias
            37. sources of error
            38. counfounding
            39. Measures of Association
            40. incidence rate
            41. relative risk
            42. Retrospective Follow-Up
            43. Study Designs
            44. Follow-Up Study Design
            45. Other Designs
            46. Examples
            47. Framingham Heart Study
            48. Nurses' Health Study
            49. Women's Health Initiative
            50. American Cancer Society Cancer Prevention Study 1
         2. Nested Case-Control Study
            1. Definition: case-control study within a cohort study
            2. Advantages
            3. efficient to conduct
            4. high comparability of cases and controls because they come from common population
         3. Case-Control
            1. Types
            2. retrospective
            3. case referent
            4. case history
            5. Case-Control Study Design
            6. Examples
            7. Adenocarcinoma of the vagina
            8. Use of proton-pump inhibitors and risk of hip/femur fracture: a population-based case-control study.
            9. 2x2 Table
            10. Purpose
            11. Retrospectively identify potential risk factors of diseases or outcomes
            12. Not designed to determine effect
            13. Establish association not causality
            14. Advantages
            15. Suited for initial, explanatory studies
            16. Study rare diseases
            17. Study a disease that occurs many years after exposure
            18. Discover possible disease etiological factors
            19. Can be completed in shorter time frame
            20. Disadvantages/Biases
            21. selection bias: subjects not representative of target population
            22. Berkson's bias: admission rates differ for comparison groups
            23. Non-response bias: non-respondents may exhibit exposures or outcomes that differ from repsonders causing over or under estimation of odds/risk
            24. Neyman bias: timing of exposure identification causes cases to be missed
            25. Unmasking bias: an innocent exposure causes a sign or symptom that precipitates search for a disease but does not itself cause the disease
            26. information bias
            27. recall bias: cases more likely to recall exposure than controls
            28. prevarication bias: lying
            29. incomparable medical records: incomplete medical records
            30. interviewer/abstractor bias: knowledge of disease state/outcome might influence intensity of a search for exposure
            31. data analysis bias -- see bias section for all observational studies
            32. sources of error -- see sources of error section for all observational studies
            33. confounding -- see confounding section for all observational studies
            34. Components
            35. Selection of Cases
            36. subjects with a particular outcome
            37. case definition
            38. use rigorous criteria
            39. separate out really severe cases
            40. use objective criteria
            41. Sources of Cases
            42. should be representative of population
            43. incident cases are best because they are new, more representative of population, and easier to identify
            44. prevalent cases give you more cases but it's harder to see relationship between exposure and outcome
            45. Selection of Controls
            46. subjects who have not experienced the outcome
            47. controls should have same characteristics as the cases except for outcome
            48. sources of controls
            49. population based
            50. institution based
            51. controls should represent larger population
            52. use of multiple controls
            53. controls of the same type, only ratio is different
            54. controls of different types
            55. used to determine a consistent association regardless of population
            56. Matching
            57. control for variables known to be related to outcome that may confuse results
            58. reduces competing explanations for outcome in question
            59. most commonly matched variables
            60. age
            61. sex
            62. race
            63. socioeconomic status
            64. Data Collected About Past Exposure
            65. Relies on people to recall past exposure and accuracy of medical records
            66. Difficult to validate
            67. Measure of Association -- Odds Ratio (OR)
            68. Statistical test used to determine association between outcome and exposure.
         4. Cross-Sectional (Prevalence Study)
            1. Types
            2. prevalence
            3. Cross-Sectional Study Design
            4. Examples
            5. Birth rates among adolescents 15 to 19 years
            6. Prevalence of cigarette smoking among U.S. adolescents from 1974 to 1991
            7. Prevalence of congenital malformations across maternal age
            8. 2x2 Table
            9. General Characterisitics
            10. Exposure and disease measures obtained at level of individual
            11. Single period of observation
            12. Quantitative estimate of magnitude of a problem
            13. Source of hypotheses for future etiologic studies
            14. Purpose
            15. Evaluate a new test or new application of an old one
            16. Evaluate predictive capability of clinical features
            17. Identify etiological agents or causative factors
            18. Determine prevalence of a problem
            19. Advantages
            20. Good for initial, explanatory studies
            21. Study rare diseases
            22. Inexpensive, quick
            23. Less involvement for px than cohort or clinical trial
            24. Can be used to identify cases and controls for case-control study
            25. Disadvantages/Biases
            26. Can't us to find out what happened first, only if they had outcome and if they had exposure
            27. Subject Selection
            28. population selection
            29. sampling selection
            30. systematic sampling
            31. random sampling
            32. Response/participation Bias: difference between those who participate and those who don't
            33. Time-order relationships: for example, does obesity or lack of activity come first
            34. Errors in data collection: how are you asking the questions
            35. Self-reported Information: are they telling the truth or do they remember everything
            36. Transient Effects: temporary occurrences that take place during the study
            37. Exclusion of cases due to rapid death or recovery might bias estimates of prevalence
            38. Risk factors from prevalent cases may differ from incident cases
      2. Problems Common to All Observational Studies
         1. Data Analysis Bias
            1. post hoc significance bias: decisions regarding significance level made posteriori
            2. data dredging bias: data analyzed for all possible associations without prior hypotheses
            3. correlation bias: correlations interpreted as causation
            4. significance bias: statistical significance is confused with clinical significance
         2. Source of Errors
            1. random errors: reflect fluctuations around a true value due to sampling variability
            2. systematic error
            3. measurement bias
            4. confounding
            5. criteria for a factor to be a confounder
            6. be a risk factor for the disease
            7. be associated with exposure under study
            8. not be an intermediate step between exposure and disease
            9. control of confounding
            10. restriction
            11. matching
            12. data analysis
            13. definition: distortion of the estimate of the effect of an exposure because it is mixed with the effect of an extraneous factor
            14. confounding variables: variables which compete with the hypothesized risk factor as an explanation for the observed response
      3. Use of 2x2 Tables to Represent Observational Studies
   2. Experimental Studies: Have Interventions
      1. Controlled Trial
         1. clinical trial
         2. educational intervention
         3. health-care trial
      2. Experimental Study Design
      3. Examples
         1. Maternal or infant antiretroviral drugs to reduce HIV-1 transmission.
2. DESCRIPTIVE
   1. Types
      1. Case Reports and Case Series
         1. definition: case report contains observations of the clinical course of a single patient and a case series of multiple patients
         2. Purpose
            1. record events, observations, activities
            2. communicate significant findings
            3. initial step in formulation of hypotheses
         3. Scope
            1. use of new therapy for some condition
            2. occurrence of specific ADR
            3. poisoning
            4. unexpected presentation of disease
         4. Advantages/Significance
            1. early recognition of drug toxicities and teratogenicity
            2. early identification of potential toxicities of dietary supplements
            3. identifying treatments for rare disorders
         5. Disadvantages
            1. no causality
      2. Clinical Series
      3. Population
      4. Program or Course
   2. 3 Important Characteristics
      1. person
      2. place
      3. time
   3. Common Uses
      1. epidemiology