Health Science midterm #2

Card Set Information

Health Science midterm #2
2013-11-06 16:13:39
Chapter 12 13

Randomized controls trials and causal inference confounding variables
Show Answers:

  1. How is the exposure given in Randomized control trials?
    investigator assigns the exposure variable
  2. What are experimental studies most commonly used for?
    testing effectiveness of health care services
  3. how are the volunteers put into the exposed and unexposed group in RCT?
    random allocation
  4. How do you analyze the data in a RCT?
    compare incidence rates in exposed (treated) and unexposed (untreated)
  5. What is equipoise? (2)
    • ethical justification of RCT
    • no convincing evidence that A is better or less toxic than B
  6. In RCT, what is the purpose of random allocation?
    to ensure similar groups
  7. Randomization in RCT , reduces ___________ error, nut __________ error may still be present
    • systematic error
    • random error
  8. What kind of bias does randomization reduce?
    selection bias
  9. Ideally, the arm to which the next participant is going to be assigned should be ________.
  10. In randomization, neither the ________ or the __________ should be able to tell any difference
    • investigator
    • patient
  11. Randomization reduces the chances that groups are different with regard to _________ and _______ risk factors.
    • known
    • unknown
  12. What is placebo?
    • medicine adapted more to please than benefit the patient
    • fake medicine
  13. What is blinding?
    when nobody knows who received what treatments
  14. what is single blind?
    only patient blinded
  15. what is double blind?
    investigators and patient blinded
  16. what is triple blind?
    investigator, patient and statisticians blinded
  17. What are the pros of RCT? (2)
    • we can infer causation
    • random allocation reduces allocation bias at baseline
  18. what are the cons to RCT? (2)
    • expensive and non-compliance
    • may be inappropriate in certain scenarios (long latency period between exposure and outcome)
  19. What relationships are RCT best for?
    best to test cause and effect relationship
  20. Koch provided a framework for identifying __________ of infectious disease.
  21. What are Koch's postulates? (3)
    • The agent has to be present in every case of the disease.
    • agent has to be isolated and grown in pure culture
    • the agent has to cause disease when inoculated into a susceptive animal and the agent must then be able to be recovered from that animal and identified
  22. What are the three reasons that Koch's postulates are useful but not rules?
    • disease production may require cofactors
    • viruses cannot be cultured like bacteria b/c viruses need living cells to grow
    • pathogenic viruses can be present without a clinical disease
  23. What is Sir Austin Bradford Hill's Causal Considerations? (9)
    • strength of association: size of RR
    • consistency: study replication
    • specificity: exposure leads to single disease
    • temporality: exposure precedes disease
    • dose-response relationship
    • plausibility: consistent with biology
    • coherence: natural history of disease
    • experimental evidence: animal or human
    • analogy
  24. What was the purpose of Hill's causal considerations?
    guidelines to help determine if associations are causal
  25. How is strength of association measured?
    by the risk ratio, relative risk
  26. How do you know the relationship between X and Y is not due to an extraneous variable?
    the stronger the relationship between the independent and dependent variable, the less likely the relationship between X and Y is not due to an extraneous variable
  27. How does consistency increase the credibility of a finding?
    multiple observations, of an association, with different people under different circumstances and with different measurement instruments increase the credibility of a finding.
  28. How does temporality help with casual consideration?
    Exposure MUST precede outcome
  29. How does coherence help with the causal considerations?
    the association must be coherent with other knowledge
  30. How does specificity help with the causal considerations?
    showing that an outcome is best predicted by one primary factor odds credibility to a causal claim
  31. How does dose-response relationship help with the causal considerations?
    There should be a direct relationship between the risk factor (independent variable) and people's status of the disease variable (dependent variable)
  32. How does plausibility help with the causal considerations?
    A hypothesized relationship, between exposure and outcome, must make sense with current biological knowledge.
  33. How does experimental evidence help with the causal considerations?
    Any related research that is based on experiments will make a causal inference more plausible
  34. What is a cause?
    is an event condition or characteristic that preceded the disease event and without which the disease event would either not have occurred at all or would have occurred later on.
  35. What is the three main criteria for a confounding variable?
    • variable associated with exposure
    • variable risk factor for outcome
    • variable does NOT fall along the causal pathway between exposure and outcome
  36. What is a confounding variable?
    A non-causal association between an exposure and an outcome is observed as a result of the influence of a third variable
  37. Why is it important to consider confounding variables?
    it can lead to make invalid conclusions
  38. What is the first step to addressing confounding? (2)
    • know your subject area
    • know which variables are likely to confound the relationship between your primary exposure and outcome
  39. How do you prevent or handle confounding in your research at the design stage? (2)
    • balance known and unknown confounders by randomizing your subjects
    • restrict your study population to one level of the confounder
  40. How do you prevent or handle confounding in your research at the analysis stage? (2)
    • stratify on the confounder
    • use multivariable analysis techniques
  41. Confounding occurs when we confuse __________ with _________.
    • association
    • causation