Public health Flashcards
calculation for relative risk (RR)
RR = a/(a+b) / c/(c+d)
absolute risk reduction (ARR)
ARR = c/(c+d) - a/(a+b)
attributable risk (AR)
AR = RR-1 / 1 X 100
(RR = a/(a+b) / c(c+d)
what does sensitivity and specificity of a test indicate
sensitive is highly sensitive when negative - rules out disease
specificity is highly specific when positive - rules in disease
how to work out true negative and true positive
TN = TN / FP
TP = TP / FN
positive predictive value calculation
PPV = TP (TP + FP)
negative predictive value calculation
NPV = TN (TN + FN)
test efficacy calculation
test efficacy = TN + TP ( TN+TP+FN+FP)
prevalence calculation
prevalence = everyone with disease / entire population
how to calculate sensitivity and specificity
sensitivity = 1-false negative rate
specificity = 1-false positive rate
what type of study identifies relative risk
cohort study
what type of study identifies odds ratio
case control
what type of study measures prevalene
cross section
what type of study measures incidence
cohort
what is best to quantify results when data is skewed
median
what is best to quantify results when date isnt skewed
mean
what phase of trial assesses side effects
phase 2
what phase of clinical trail assesses pharmacokinestics
phase 0
what phase of clinical trial assesses dosing
phase 1
what are the 5 phases of clinical trial
phase 0 - Initial pharmacokinetics
phase 1 - Safety ? dosing
phase 2 - will it work? adverse effects
phase 3 - Improvements? compares placebo
phase 4 - stay on the Market? long term/rare side effects
what type of clinical trial uses their own subjects as controls
cross-over clinical trial
compares > 2 treatments in a group of patients by having a washout station
calculation for number needed to treat (NNT)
NNT = the number of people needed to treat to have 1 patient show benefit
NNT = 1/ARR
(ARR = c/c+b - a(a+b)
what does a poor test-retest reliability indicate
indicates a low ability to reproduce a result = low precision
