Ethics & Stats & & ICD Codes, Oh My Flashcards
mishmash (17 cards)
describe informed refusal
decision against recommeded medical advise
describe proxy consent
given by legally appointed health care proxy
describe surrogate decision
decision made by family member or clos friend (not legally appointed)
describe therapeutic priviledge
limited physician privilege of doc to withhold info from patient if info will seriously harm patient
stats that are affected by prevalence
positive and negative predictive value
sensitivity, specificity do not change
as prevalence decreases, PPV decreases, NPV incrases
describe sensitivitiy
describe high sensitivity
TEST’s ability to correctly identify pt w/ d/z
high true positive rate
also leads to false positives (test incorrectly says pt has d/z when they dont)
describe specificity
describe high specificity
TEST’s ability to correctly identify those without d/z
high true negatives (minimize false positives)
also leads to false negatives (test incorrectly says pt doesn’t have d/z, when they do)
sensitivity calculation
TP / (TP+FN)
identified cases / (all cases of d/z)
specificity calculation
TN / (TN + FP)
identified non-sick / (all non-sick cases)
describe TYPE I error
wolf analogy
You reject a null hypothesis when the null hypothesis is true
reject the null hypothesis (no wolf) and say there is (boy cry wolf)
describe TYPE II error
wolf analogy
You accept the null hypothesis when the null hypothesis is false
accept null (village doesn’t listen) when shouldnt (there is a wolf!)
odds ratio reported in what kind of study?
why
case-control
case-control can’t produce incidence of d/z
measure of associatoin
relative risks reported in what kind of study?
cohort study
measure of associatoin
OR calculation
and why?
ad/cb
“cross products”
divide odds in Group 1 (case) by odds in Group 2 (control)
(a/c) / (b/d)
odds that a case was exposed = (a/c)
odds that a control was exposed = (b/d)
describe justice
refers to patient equality and fairness
has to do with resource allocation, not money? confusing
best study for rare d/z
case series
Join Commission MC reported sentinel event
falls
