research Flashcards
formula for sensitivity
TP/(TP+FN)
testing for true positive
good screener- good at ruling out
formula for specificity
TN/(TN+FP)
testing for true negative
good at confirming- rule in
formula for PPV
TP/(TP+FP)
someone having the disease actually testing positive for it
formula for NPV
TN/(TN+FN)
someone healthy actually testing negative for it
NNT formula
1/absolute risk reduction
how is PPV and NPV related to prevalence
greater the prevalence the higher the PPV
less the prevalence the higher the NPV
at what p value does it indicate a statistically significant result
p < 0.05 is statistically significant to reject the null hypothesis
this means there is a 5% likelihood that your results are just d/t chance
what is a type I error
rejecting the null hypothesis in error
same as the P value
what is a type II error
accepting the null hypothesis in error
what is validity
whether and instrument or test actually measured what it is support to measure
think ear thermometer
what is reliability
the consistency of reults
intention to treat
analysis that insists that all pts remain in the original groups they were initially randomly assigned in the study
helps increase the power of a study
what is NNT
number of pts you would need to tx to prevent one additional adverse affect
what is incidence
number of newly dx cases
what is prevalence
number of total cases in a population
what is standard error
the accuracy of the sample mean in comparison to the true population mean
for example— determine average height in 10 y/o girls ned a sample with rage of heights that are representative of the general population
what is pre test probability
the estimate of the probability of a condition present before you do testing
think about criteria/guidelines before testing is done (ie centaurs criteria before rapid strep)
likelihood ratio
a person with the disease will have a positive test result
LR greater than 1 increases the probability that the target condition is present
LR less than 1 makes the condition less likely
formula for absolute risk
a/(a+b)
a= exposed + disease b= exposed + healthy
formula for relative risk
(a/(a+b))/(c/(c+d))
a= exposed + disease b= exposed + healthy c= not exposed + disease d= not exposed + healthy
formula for odds ratio
(a/(a+b))/(b/(a+b)) or a/b
a= exposed + disease b= exposed + healthy
what is the best type of scientific study
RCT
gold standard of research design
what is a cohort study?
cohort is followed forward or traced backwards to investigate associated risks for a certain outcome
need a large sample size
what is the difference in case control and cohort studies
cohort uses a large sample size to evaluated associated risks and outcome
case control uses a small sample size of individuals with disease (case) compared to a control and then check back for risk factors
