Cancer Risk Flashcards
absolute risk
incidence of a disease in a population (cases per total people)
relative risk
absolute risk in unexposed relative to exposed (case/persons exposed)/(cases/persons unexposed)
odds ratio
odds of diseased person having had exposure relative to odds of non-diseased having had exposure (1/2 diseased) / (1/10 nondiseased)
OR accurately estimates
Relative Risk
Measures of association
relative risk and odds ratio
cohort studies
relative risk
RR =
a/(a+b) / c/(c+d)
case control studies
odds ration
OR =
a/c / b/d (cancer/cancer / no/no)
Excess Relative Risk (ERR)
RR - 1
Ecological Studies
association with disease and exposure
LInear model
might overestimate risk
Quadratic model
might underestimate risk
Unacceptable models of risk
threshold
hormesis
attributable risk
Risk in exposed - risk in unexposed
