Lecture #6: Life Tables

Lecture #6: Life Tables - Review of Study Designs and Life...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
Review of Study Designs and Life Tables 7 May 2008 Session 12
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Experimental Studies: The Gold Standard Assignment of exposure is RANDOM (receiving the exposure is completely due to chance); follow-up over time to compare incidence in outcome between the different exposure groups. Randomized controlled trials (individual, clinical trials) e.g. to determine the impact of acupuncture on lower back pain, randomly assign each participant to either acupuncture or sham (fake) acupuncture, and follow-up over time to compare improvement in lower back pain. Community intervention (unit of analysis is the community, not the individual!) e.g. to determine the impact of providing a new dengue (a very painful disease caused by a viral agent with a mosquito vector) education campaign in an Amazonian community, 20 villages are randomized to either standard education or the ramped up education campaign and followed up over time for dengue incidence.
Background image of page 2
Observational Studies: Cohort Study Assignment of exposure is NOT random (systematic differences between exposed and unexposed persons possible!) Look at exposed and unexposed subjects separately and follow-up over time to compare incidence in outcome between the different exposure groups. E.g. You want to determine if infants who are breastfed have a lower incidence of diarrhea compared to those who aren’t in a developing country. You follow those who are breastfed and those who are not, and compare their incidence of diarrhea over a 1 year period.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Observational Studies: Case- Control Study Look at diseased and non-diseased subjects separately and compare their exposures from the past. E.g. Among the elderly, you want to determine if past lifestyle (say, back when they were in their 30’s and 40’s) is associated with current cardiovascular disease (CVD). You look at those who have CVD vs. those who don’t, and compare their lifestyle back when they were young (e.g. past diet, exercise patterns).
Background image of page 4
Observational Studies: Cross- Sectional Study Assess exposure and outcome at same point in time; direction of causality ( temporality ) can be a big problem! E.g. A one time survey among middle school students finds that kids who play >3 hours of video games daily are overweight compared to those who play <1 hour a day. It could be that playing videogames for long hours leads to weight gain, OR those who are already overweight tend to be less active and prefer to play videogames sitting down. Temporality is unclear!
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Observational Studies: Case- Series Assess multiple cases to describe patterns among them. Like case-control, but no comparison group (no controls)! Causal inference can be very limited. E.g. The first AIDS cases in the US was among 5
Background image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 49

Lecture #6: Life Tables - Review of Study Designs and Life...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online