evidence based medicine Flashcards
(19 cards)
best study to determine prevalence of a disease
cross sectional study
a study that compares 2 groups of 4 y/o with similar characteristics: one group is given a drug + the other a placebo, growth of each is measured after this intervention - what kind of study?
controlled trial
a study that aims to establish the noraml height of 4y/o by measuring heights at school entry
cross sectional
screening
a process of idenitifying apparently healthy people who may be at increased risk of disease or condition
(normal/negative result on screening does not equal disease free necessarily)
sensitivity
how well the test detects having the disease
no of results where test is positive in those with disease / numbr of people with the disease
specificity
how well test detects NOT having the disease
no of “normal” results where disease in NOT detected in people WITHOUT the disease / no of people without disease
positive predictive value
how reliable is the test result when it shows disease is present
no of people with a positive test + have disease / number of people with a pos test
neg predicitve value
how reliable is the test result when disease NOT present
no of people who have a neg test result + do NOT have disease / no of people with a neg test result
effect of prevalence on positive + negative predictive values
As prevalence increases -> PPV also increases but the NPV decreases
prevalence decreases -> PPV decreases while the NPV increases
a study that compares the height of a group of 4y/o living near a nuclear plant with the height of a group of 4y/o who live elsewhere
case control studies
controlled trial vs case control studies
controlled - one group given drug the other placebo + compares (you’re adding something)
case control - comparing outcomes of people in different locations (environment all ready exists)
kids born at one hospital 1 year + measure their height at intervals of up to 4yrs of age
cohort study
external validity
the extent to whoch one can appropriately apply the results to other popultions
OTHER -> external validiation
which study design is the best to research the aetiology of a disease?
cohort
-> time frame important
define statistical significance
the results of a study are unlikely to have arisen by chance alone
what does qualitative research seek to analyse?
the date which reserachers collect from people in relation to their experiences + circumstances
saturation
no longer any need to sample more people to reach new conclusion or to back up or challenge existing conclusions
absolute risk (AR)
the number of events in treated or control groups, divided by the number of people in that group
ARC = AR of events in the control group
ART = AR of events in the treatment group
ARR(absolute risk reduction) = ARC - ART
RR ( relative risk) = ART / ARC
how is the number needed to treat calculated?
NNT = 1/ARR –> (1 in …)
ARR (absolute risk reduction) = ARC - ART
