Biostats/Epi Flashcards
Number needed to treat, Number needed to harm
NNT = 1/(ARR) = 1/(control rate - treatment rate). ARR = absolute relative risk. Lower NNT = more beneficial treatment
NNH = 1/(ARI) = 1/(treatment rate - control rate). ARI = absolute risk increase
Odds ratio vs. relative risk (formulas and when to use)
OR used for case control studies; OR = ad/bc using standard contingency table
RR used for cohort studies; RR = (a/(a+b))/(c(c+d))
Absolute risk reduction (ARR )
ARR = (control rate) - (tx rate)
Relative risk reduction (RRR)
RRR = ARR/control rate = 1 - RR RRR = (control rate - tx rate)/control rate
Phases of clinical trials
“can the drugs SWIM”?
Phase 1: is it Safe? (safety, tox, pharmacokinetics/dynamics)
Phase 2: does it Work? (efficacy, optimal dosing, adverse fx)
Phase 3: any Improvement? (compare to current std of care)
Phase 4: can it stay on the Market? (detect rare/long-term adverse fx)
With what values of a confidence interval would you not reject H0 (not statistically significant)?
If 95% CI for mean diff btwn 2 variables includes 0, or if 95% CI for RR or OR includes 1, there is no sig diff and H0 is not rejected
Attributable risk percent (ARP)
ARP = (RR - 1)/RR ARP = (risk in exposed - risk in unexposed)/(risk in exposed)
Crossover study
Study in which subjects act as their own controls. Reduces confounding (just like matched studies, or multiple/repeated studies)