EPID600 (Fall 2009) module on
Data analysis and interpretation / causal inference


Instructions:
  1. Read: Aschengrau and Seage, ch. 15 - The Epidemiologic Approach to Causation

  2. Read the lecture slides and attend the lecture (if you cannot attend the lecture, the slides have verbatim speaker notes; there are also recorded audio tutorials, but these are not as up-to-date as the speaker notes).

  3. Read Matthias Egger, George Davey Smith. Meta-analysis: potentials and promise. British Medical Journal 22 Nov 1997;315:1371-1374. (UNC-CH: full text) - scroll down to page 1371.)
    Supplemental resources can be found at www.unc.edu/epid600/modules/120-DataAnalysis/

  4. Read :
    Stephen Moses, Francis A. Plummer, Janet E. Bradley, Jeckoniah O. Ndinya-Achola, Nico J.D. Nagelkerke, Allan R. Ronald. The association between lack of male circumcision and risk for HIV infection: a review of the epidemiological data. Sexually Transmitted Diseases, July-August 1994;21(4):201-210. (UNC-CH: full text))
        and
    Jayne Byakika-Tusiime. Circumcision and HIV infection: assessment of causality. AIDS and Behavior 2008;12:835-841. (abstract, UNC-CH: HTML full text), UNC-CH: PDF full text)

  5. Answer (all of) the case study questions. Write succinct answers that include a brief explanation or statement of support.

  6. (Optional, but earns credit) Submit the answers to the starred case study questions (numbers 1, 2, 3, 7, and 11 - see the class schedule for due date).

gc. 2009cd