biostats Flashcards Preview

Boards: Pharm and Biostats > biostats > Flashcards

Flashcards in biostats Deck (14):

What kind of stats come from a case control study?

odds ratio: " patients with COPD had a higher odds of a history of smoking than those without COPD had"


what kind of stats come from a cohort study

relative risk. compares a group with a given exposure to a group without such an exposure. "smokers had a higher risk of developing COPD than non-smokers."


What questions are asked at each phase of a clinical trial?

phase I: is it safe?
phase II: Does it work? (efficacy, dosing, adverse effects)
phase III: Is it as good or better than other treatments?
phase IV: postmarketing surveillence of pts after approval- looks lfor rare/long-term effects of the drug


when is high sensitivity the most important?

when you are "ruling out" a condition.
also needed when screening a disease with low prevalence


odds ratio

used in case-control studies
odds that the group with the disease was exposed to a risk factor divided by the odds that the group without the disease was exposed. (a/c)/(b/d)


relative risk

used in cohort studies. risk of developing the disease in the exposed group divided by the risk in the unexposed group. (a/(a+b))/(c/(c+d))


relative risk reduction

proportion of risk reduction attributable to the intervention: RRR = 1-RR. (if 2% of pts who receive a flu shot develop flu vs. 8% of unvaccinated, RR = 2/8 = .25. RRR = .75)


attributable risk

difference in risk btw exposed and unexposed groups, or the proportion of disease occurrences that are attributable to the exposure. (a/(a+b))-(c/(c+d))


absolute risk

difference in risk attributable to the intervetion as compared to a control. (8% of ppl who receive a placebo get flu vs 2% who receive a flu shot, then 8%-2%= 0.06


number needed to treat

1/absolute risk reduction


number needed to harm

number of pts who need to be exposed to a risk factor for 1 pt to be harmed. 1/absolute risk


Hawthorne effect

groups who know they're being studied behave differently than they would otherwise


How are crossover studies useful?

can reduce confounding bias. subjects act as their own controls (they switch from exposure to non-exposure), but obviously keep the confounding variables the same
matching can also reduce confounding


How do you address lead time bias?

measure back end survival- adjust survival according to the severity of disease at diagnosis